mirror of
https://github.com/openai/whisper.git
synced 2025-11-24 06:26:03 +00:00
Merge 1f8e215ed52ea01503fe7de5efc1f16287977173 into c0d2f624c09dc18e709e37c2ad90c039a4eb72a2
This commit is contained in:
Arya vaghayenegar
GitHub
commit
ac2691c67f
28
.railwayignore
Normal file
28
.railwayignore
Normal file
@ -0,0 +1,28 @@
|
|||||||
|
# Railway ignore file - prevent large files from being included in build
|
||||||
|
node_modules/.bin
|
||||||
|
node_modules/.cache
|
||||||
|
node_modules/.vite
|
||||||
|
.next
|
||||||
|
.nuxt
|
||||||
|
dist
|
||||||
|
build
|
||||||
|
*.log
|
||||||
|
.DS_Store
|
||||||
|
.env.local
|
||||||
|
.venv
|
||||||
|
venv
|
||||||
|
__pycache__
|
||||||
|
*.pyc
|
||||||
|
.pytest_cache
|
||||||
|
.coverage
|
||||||
|
.git
|
||||||
|
.github
|
||||||
|
notebooks
|
||||||
|
tests
|
||||||
|
data
|
||||||
|
*.png
|
||||||
|
*.svg
|
||||||
|
.flake8
|
||||||
|
.pre-commit-config.yaml
|
||||||
|
CHANGELOG.md
|
||||||
|
model-card.md
|
||||||
305
QUICKSTART.md
Normal file
305
QUICKSTART.md
Normal file
@ -0,0 +1,305 @@
|
|||||||
|
# Farsi Transcriber - Quick Start Guide
|
||||||
|
|
||||||
|
You now have **TWO** complete applications for Farsi transcription:
|
||||||
|
|
||||||
|
## 🖥️ Option 1: Desktop App (PyQt6)
|
||||||
|
|
||||||
|
**Location:** `/home/user/whisper/farsi_transcriber/`
|
||||||
|
|
||||||
|
### Setup
|
||||||
|
```bash
|
||||||
|
cd farsi_transcriber
|
||||||
|
python3 -m venv venv
|
||||||
|
source venv/bin/activate
|
||||||
|
pip install -r requirements.txt
|
||||||
|
python main.py
|
||||||
|
```
|
||||||
|
|
||||||
|
**Features:**
|
||||||
|
- ✅ Standalone desktop application
|
||||||
|
- ✅ Works completely offline
|
||||||
|
- ✅ Direct access to file system
|
||||||
|
- ✅ Lightweight and fast
|
||||||
|
- ⚠️ Simpler UI (green theme)
|
||||||
|
|
||||||
|
**Good for:**
|
||||||
|
- Local-only transcription
|
||||||
|
- Users who prefer desktop apps
|
||||||
|
- Offline processing
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 🌐 Option 2: Web App (React + Flask)
|
||||||
|
|
||||||
|
**Location:** `/home/user/whisper/farsi_transcriber_web/`
|
||||||
|
|
||||||
|
### Setup
|
||||||
|
|
||||||
|
**Backend (Flask):**
|
||||||
|
```bash
|
||||||
|
cd farsi_transcriber_web/backend
|
||||||
|
python3 -m venv venv
|
||||||
|
source venv/bin/activate
|
||||||
|
pip install -r requirements.txt
|
||||||
|
python app.py
|
||||||
|
# API runs on http://localhost:5000
|
||||||
|
```
|
||||||
|
|
||||||
|
**Frontend (React):**
|
||||||
|
```bash
|
||||||
|
cd farsi_transcriber_web
|
||||||
|
npm install
|
||||||
|
npm run dev
|
||||||
|
# App runs on http://localhost:3000
|
||||||
|
```
|
||||||
|
|
||||||
|
**Features:**
|
||||||
|
- ✅ Modern web-based UI (matches your Figma design exactly)
|
||||||
|
- ✅ File queue management
|
||||||
|
- ✅ Dark/Light theme toggle
|
||||||
|
- ✅ Search with text highlighting
|
||||||
|
- ✅ Copy segments to clipboard
|
||||||
|
- ✅ Resizable window
|
||||||
|
- ✅ RTL support for Farsi
|
||||||
|
- ✅ Multiple export formats
|
||||||
|
- ✅ Professional styling
|
||||||
|
|
||||||
|
**Good for:**
|
||||||
|
- Modern web experience
|
||||||
|
- Team collaboration (can be deployed online)
|
||||||
|
- More features and polish
|
||||||
|
- Professional appearance
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 📊 Comparison
|
||||||
|
|
||||||
|
| Feature | Desktop (PyQt6) | Web (React) |
|
||||||
|
|---------|-----------------|------------|
|
||||||
|
| **Interface** | Simple, green | Modern, professional |
|
||||||
|
| **Dark Mode** | ❌ | ✅ |
|
||||||
|
| **File Queue** | ❌ | ✅ |
|
||||||
|
| **Search** | ❌ | ✅ |
|
||||||
|
| **Copy Segments** | ❌ | ✅ |
|
||||||
|
| **Resizable Window** | ❌ | ✅ |
|
||||||
|
| **Export Formats** | SRT, TXT, VTT, JSON, TSV | TXT, SRT, VTT, JSON |
|
||||||
|
| **Offline** | ✅ | Requires backend |
|
||||||
|
| **Easy Setup** | ✅✅ | ✅ (2 terminals) |
|
||||||
|
| **Deployment** | Desktop only | Can host online |
|
||||||
|
| **Code Size** | ~25KB | ~200KB |
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 🚀 Which Should You Use?
|
||||||
|
|
||||||
|
### Use **Desktop App** if:
|
||||||
|
- You want simple, quick setup
|
||||||
|
- You never share transcriptions
|
||||||
|
- You prefer offline processing
|
||||||
|
- You don't need advanced features
|
||||||
|
|
||||||
|
### Use **Web App** if:
|
||||||
|
- You like modern interfaces
|
||||||
|
- You want dark/light themes
|
||||||
|
- You need file queue management
|
||||||
|
- You want to potentially share online
|
||||||
|
- You want professional appearance
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 📁 Project Structure
|
||||||
|
|
||||||
|
```
|
||||||
|
whisper/
|
||||||
|
├── farsi_transcriber/ (Desktop PyQt6 App)
|
||||||
|
│ ├── ui/
|
||||||
|
│ ├── models/
|
||||||
|
│ ├── utils/
|
||||||
|
│ ├── config.py
|
||||||
|
│ ├── main.py
|
||||||
|
│ └── requirements.txt
|
||||||
|
│
|
||||||
|
└── farsi_transcriber_web/ (Web React App)
|
||||||
|
├── src/
|
||||||
|
│ ├── App.tsx
|
||||||
|
│ ├── components/
|
||||||
|
│ └── main.tsx
|
||||||
|
├── backend/
|
||||||
|
│ ├── app.py
|
||||||
|
│ └── requirements.txt
|
||||||
|
├── package.json
|
||||||
|
└── vite.config.ts
|
||||||
|
```
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 🔧 System Requirements
|
||||||
|
|
||||||
|
### Desktop App
|
||||||
|
- Python 3.8+
|
||||||
|
- ffmpeg
|
||||||
|
- 4GB RAM
|
||||||
|
|
||||||
|
### Web App
|
||||||
|
- Python 3.8+ (backend)
|
||||||
|
- Node.js 16+ (frontend)
|
||||||
|
- ffmpeg
|
||||||
|
- 4GB RAM
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 📝 Setup Checklist
|
||||||
|
|
||||||
|
### Initial Setup (One-time)
|
||||||
|
|
||||||
|
- [ ] Install ffmpeg
|
||||||
|
```bash
|
||||||
|
# Ubuntu/Debian
|
||||||
|
sudo apt install ffmpeg
|
||||||
|
|
||||||
|
# macOS
|
||||||
|
brew install ffmpeg
|
||||||
|
|
||||||
|
# Windows
|
||||||
|
choco install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
- [ ] Verify Python 3.8+
|
||||||
|
```bash
|
||||||
|
python3 --version
|
||||||
|
```
|
||||||
|
|
||||||
|
- [ ] Verify Node.js 16+ (for web app only)
|
||||||
|
```bash
|
||||||
|
node --version
|
||||||
|
```
|
||||||
|
|
||||||
|
### Desktop App Setup
|
||||||
|
|
||||||
|
- [ ] Create virtual environment
|
||||||
|
- [ ] Install requirements
|
||||||
|
- [ ] Run app
|
||||||
|
|
||||||
|
### Web App Setup
|
||||||
|
|
||||||
|
**Backend:**
|
||||||
|
- [ ] Create virtual environment
|
||||||
|
- [ ] Install requirements
|
||||||
|
- [ ] Run Flask server
|
||||||
|
|
||||||
|
**Frontend:**
|
||||||
|
- [ ] Install Node dependencies
|
||||||
|
- [ ] Run dev server
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 🎯 Quick Start (Fastest)
|
||||||
|
|
||||||
|
### Desktop (30 seconds)
|
||||||
|
```bash
|
||||||
|
cd whisper/farsi_transcriber
|
||||||
|
python3 -m venv venv && source venv/bin/activate
|
||||||
|
pip install -r requirements.txt && python main.py
|
||||||
|
```
|
||||||
|
|
||||||
|
### Web (2 minutes)
|
||||||
|
Terminal 1:
|
||||||
|
```bash
|
||||||
|
cd whisper/farsi_transcriber_web/backend
|
||||||
|
python3 -m venv venv && source venv/bin/activate
|
||||||
|
pip install -r requirements.txt && python app.py
|
||||||
|
```
|
||||||
|
|
||||||
|
Terminal 2:
|
||||||
|
```bash
|
||||||
|
cd whisper/farsi_transcriber_web
|
||||||
|
npm install && npm run dev
|
||||||
|
```
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 🐛 Troubleshooting
|
||||||
|
|
||||||
|
### "ffmpeg not found"
|
||||||
|
Install ffmpeg (see requirements above)
|
||||||
|
|
||||||
|
### "ModuleNotFoundError" (Python)
|
||||||
|
```bash
|
||||||
|
# Ensure virtual environment is activated
|
||||||
|
source venv/bin/activate # Linux/Mac
|
||||||
|
# or
|
||||||
|
venv\Scripts\activate # Windows
|
||||||
|
```
|
||||||
|
|
||||||
|
### "npm: command not found"
|
||||||
|
Install Node.js from https://nodejs.org
|
||||||
|
|
||||||
|
### App runs slow
|
||||||
|
- Use GPU: Install CUDA
|
||||||
|
- Reduce model size: change to 'small' or 'tiny'
|
||||||
|
- Close other applications
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 📚 Full Documentation
|
||||||
|
|
||||||
|
- **Desktop App:** `farsi_transcriber/README.md`
|
||||||
|
- **Web App:** `farsi_transcriber_web/README.md`
|
||||||
|
- **API Docs:** `farsi_transcriber_web/README.md` (Endpoints section)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 🎓 What Was Built
|
||||||
|
|
||||||
|
### Desktop Application (PyQt6)
|
||||||
|
✅ File picker for audio/video
|
||||||
|
✅ Whisper integration with word-level timestamps
|
||||||
|
✅ 5 export formats (TXT, SRT, VTT, JSON, TSV)
|
||||||
|
✅ Professional styling
|
||||||
|
✅ Progress indicators
|
||||||
|
✅ Threading to prevent UI freezing
|
||||||
|
|
||||||
|
### Web Application (React + Flask)
|
||||||
|
✅ Complete Figma design implementation
|
||||||
|
✅ File queue management
|
||||||
|
✅ Dark/light theme
|
||||||
|
✅ Search with highlighting
|
||||||
|
✅ Segment management
|
||||||
|
✅ Resizable window
|
||||||
|
✅ RTL support
|
||||||
|
✅ Flask backend with Whisper integration
|
||||||
|
✅ 4 export formats
|
||||||
|
✅ Real file upload handling
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 🚀 Next Steps
|
||||||
|
|
||||||
|
1. **Choose your app** (Desktop or Web)
|
||||||
|
2. **Install ffmpeg** if not already installed
|
||||||
|
3. **Follow the setup instructions** above
|
||||||
|
4. **Test with a Farsi audio file**
|
||||||
|
5. **Export in your preferred format**
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 💡 Tips
|
||||||
|
|
||||||
|
- **First transcription is slow** (downloads 769MB model)
|
||||||
|
- **Use larger models** (medium/large) for better accuracy
|
||||||
|
- **Use smaller models** (tiny/base) for speed
|
||||||
|
- **GPU significantly speeds up** transcription
|
||||||
|
- **Both apps work offline** (after initial model download)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## 📧 Need Help?
|
||||||
|
|
||||||
|
- Check the full README in each app's directory
|
||||||
|
- Verify all requirements are installed
|
||||||
|
- Check browser console (web app) or Python output (desktop)
|
||||||
|
- Ensure ffmpeg is in your PATH
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
**Enjoy your Farsi transcription apps!** 🎉
|
||||||
114
RAILWAY_QUICKSTART.md
Normal file
114
RAILWAY_QUICKSTART.md
Normal file
@ -0,0 +1,114 @@
|
|||||||
|
# Railway Deployment - Quick Start (5 Minutes)
|
||||||
|
|
||||||
|
Deploy your Farsi Transcriber to Railway in just 5 minutes! 🚀
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **What You'll Get**
|
||||||
|
|
||||||
|
✅ Your app live online
|
||||||
|
✅ Free $5/month credit
|
||||||
|
✅ 24/7 uptime
|
||||||
|
✅ Automatic scaling
|
||||||
|
✅ No credit card needed (free tier)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Step 1: Create Railway Account** (2 min)
|
||||||
|
|
||||||
|
1. Go to **https://railway.app**
|
||||||
|
2. Click **"Login with GitHub"**
|
||||||
|
3. Authorize with your GitHub account
|
||||||
|
4. Done! You get $5 free credit ✅
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Step 2: Create Backend Service** (2 min)
|
||||||
|
|
||||||
|
1. Click **"Create New Project"**
|
||||||
|
2. Select **"GitHub Repo"**
|
||||||
|
3. Find your **whisper** fork
|
||||||
|
4. Railway auto-detects Python project
|
||||||
|
5. **Root Directory:** `farsi_transcriber_web/backend`
|
||||||
|
6. Click **Deploy**
|
||||||
|
7. **Wait 2-3 minutes** for deployment
|
||||||
|
8. **Copy the URL** that appears (e.g., `https://farsi-api-xxx.railway.app`)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Step 3: Create Frontend Service** (1 min)
|
||||||
|
|
||||||
|
1. In Railway project, click **"New Service"** → **"GitHub Repo"**
|
||||||
|
2. Select **whisper** again
|
||||||
|
3. **Root Directory:** `farsi_transcriber_web`
|
||||||
|
4. Click **Deploy**
|
||||||
|
5. **Wait 3-5 minutes** for build and deployment
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Step 4: Connect Frontend to Backend** (Bonus step - 1 min)
|
||||||
|
|
||||||
|
1. In Railway, select **frontend** service
|
||||||
|
2. Go to **Variables**
|
||||||
|
3. Edit `VITE_API_URL` and paste your backend URL from Step 2
|
||||||
|
4. Click **Deploy** to redeploy with correct API URL
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **That's It! 🎉**
|
||||||
|
|
||||||
|
Your app is now live! Click the frontend service to see your live URL.
|
||||||
|
|
||||||
|
Example URLs:
|
||||||
|
- Frontend: `https://farsi-transcriber-prod.railway.app`
|
||||||
|
- Backend: `https://farsi-api-prod.railway.app`
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Test Your App**
|
||||||
|
|
||||||
|
1. Click your frontend URL
|
||||||
|
2. Add a file
|
||||||
|
3. Click Transcribe
|
||||||
|
4. Wait for transcription
|
||||||
|
5. Export results
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Detailed Guide**
|
||||||
|
|
||||||
|
For more details, see: `farsi_transcriber_web/RAILWAY_DEPLOYMENT.md`
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Cost**
|
||||||
|
|
||||||
|
- **First 3 months:** FREE ($5/month credit)
|
||||||
|
- **After that:** ~$2-3/month for personal use
|
||||||
|
- Can upgrade to paid tier for more resources
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Common Issues**
|
||||||
|
|
||||||
|
**"API connection failed"**
|
||||||
|
- Make sure backend URL is correct in frontend variables
|
||||||
|
- Redeploy frontend after updating API URL
|
||||||
|
|
||||||
|
**"Model not loaded"**
|
||||||
|
- Wait 1-2 minutes on first transcription
|
||||||
|
- Model downloads on first use
|
||||||
|
|
||||||
|
**"Build failed"**
|
||||||
|
- Check Railway logs for errors
|
||||||
|
- Ensure all files are committed
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## **Support**
|
||||||
|
|
||||||
|
For detailed setup help, see: `farsi_transcriber_web/RAILWAY_DEPLOYMENT.md`
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
**Your Farsi Transcriber is now online!** Share the URL with anyone! 🌐
|
||||||
52
farsi_transcriber/.gitignore
vendored
Normal file
52
farsi_transcriber/.gitignore
vendored
Normal file
@ -0,0 +1,52 @@
|
|||||||
|
# Python
|
||||||
|
__pycache__/
|
||||||
|
*.py[cod]
|
||||||
|
*$py.class
|
||||||
|
*.so
|
||||||
|
.Python
|
||||||
|
build/
|
||||||
|
develop-eggs/
|
||||||
|
dist/
|
||||||
|
downloads/
|
||||||
|
eggs/
|
||||||
|
.eggs/
|
||||||
|
lib/
|
||||||
|
lib64/
|
||||||
|
parts/
|
||||||
|
sdist/
|
||||||
|
var/
|
||||||
|
wheels/
|
||||||
|
*.egg-info/
|
||||||
|
.installed.cfg
|
||||||
|
*.egg
|
||||||
|
|
||||||
|
# Virtual Environment
|
||||||
|
venv/
|
||||||
|
ENV/
|
||||||
|
env/
|
||||||
|
|
||||||
|
# IDE
|
||||||
|
.vscode/
|
||||||
|
.idea/
|
||||||
|
*.swp
|
||||||
|
*.swo
|
||||||
|
*~
|
||||||
|
.DS_Store
|
||||||
|
|
||||||
|
# PyTorch/ML Models
|
||||||
|
*.pt
|
||||||
|
*.pth
|
||||||
|
models/downloaded/
|
||||||
|
|
||||||
|
# Whisper models cache
|
||||||
|
~/.cache/whisper/
|
||||||
|
|
||||||
|
# Application outputs
|
||||||
|
transcriptions/
|
||||||
|
exports/
|
||||||
|
*.log
|
||||||
|
|
||||||
|
# Testing
|
||||||
|
.pytest_cache/
|
||||||
|
.coverage
|
||||||
|
htmlcov/
|
||||||
257
farsi_transcriber/README.md
Normal file
257
farsi_transcriber/README.md
Normal file
@ -0,0 +1,257 @@
|
|||||||
|
# Farsi Transcriber
|
||||||
|
|
||||||
|
A professional desktop application for transcribing Farsi audio and video files using OpenAI's Whisper model.
|
||||||
|
|
||||||
|
## Features
|
||||||
|
|
||||||
|
✨ **Core Features**
|
||||||
|
- 🎙️ Transcribe audio files (MP3, WAV, M4A, FLAC, OGG, AAC, WMA)
|
||||||
|
- 🎬 Extract audio from video files (MP4, MKV, MOV, WebM, AVI, FLV, WMV)
|
||||||
|
- 🇮🇷 High-accuracy Farsi/Persian language transcription
|
||||||
|
- ⏱️ Word-level timestamps for precise timing
|
||||||
|
- 📤 Export to multiple formats (TXT, SRT, VTT, JSON, TSV)
|
||||||
|
- 💻 Clean, intuitive PyQt6-based GUI
|
||||||
|
- 🚀 GPU acceleration support (CUDA) with automatic fallback to CPU
|
||||||
|
- 🔄 Progress indicators and real-time status updates
|
||||||
|
|
||||||
|
## System Requirements
|
||||||
|
|
||||||
|
**Minimum:**
|
||||||
|
- Python 3.8 or higher
|
||||||
|
- 4GB RAM
|
||||||
|
- ffmpeg installed
|
||||||
|
|
||||||
|
**Recommended:**
|
||||||
|
- Python 3.10+
|
||||||
|
- 8GB+ RAM
|
||||||
|
- NVIDIA GPU with CUDA support (optional but faster)
|
||||||
|
- SSD for better performance
|
||||||
|
|
||||||
|
## Installation
|
||||||
|
|
||||||
|
### Step 1: Install ffmpeg
|
||||||
|
|
||||||
|
Choose your operating system:
|
||||||
|
|
||||||
|
**Ubuntu/Debian:**
|
||||||
|
```bash
|
||||||
|
sudo apt update && sudo apt install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
**Fedora/CentOS:**
|
||||||
|
```bash
|
||||||
|
sudo dnf install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
**macOS (Homebrew):**
|
||||||
|
```bash
|
||||||
|
brew install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
**Windows (Chocolatey):**
|
||||||
|
```bash
|
||||||
|
choco install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
**Windows (Scoop):**
|
||||||
|
```bash
|
||||||
|
scoop install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
### Step 2: Set up Python environment
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# Navigate to the repository
|
||||||
|
cd whisper/farsi_transcriber
|
||||||
|
|
||||||
|
# Create virtual environment
|
||||||
|
python3 -m venv venv
|
||||||
|
|
||||||
|
# Activate virtual environment
|
||||||
|
source venv/bin/activate # On Windows: venv\Scripts\activate
|
||||||
|
```
|
||||||
|
|
||||||
|
### Step 3: Install dependencies
|
||||||
|
|
||||||
|
```bash
|
||||||
|
pip install -r requirements.txt
|
||||||
|
```
|
||||||
|
|
||||||
|
This will install:
|
||||||
|
- PyQt6 (GUI framework)
|
||||||
|
- openai-whisper (transcription engine)
|
||||||
|
- PyTorch (deep learning framework)
|
||||||
|
- NumPy, tiktoken, tqdm (supporting libraries)
|
||||||
|
|
||||||
|
## Usage
|
||||||
|
|
||||||
|
### Running the Application
|
||||||
|
|
||||||
|
```bash
|
||||||
|
python main.py
|
||||||
|
```
|
||||||
|
|
||||||
|
### Step-by-Step Guide
|
||||||
|
|
||||||
|
1. **Launch the app** - Run `python main.py`
|
||||||
|
2. **Select a file** - Click "Select File" button to choose audio/video
|
||||||
|
3. **Transcribe** - Click "Transcribe" and wait for completion
|
||||||
|
4. **View results** - See transcription with timestamps
|
||||||
|
5. **Export** - Click "Export Results" to save in your preferred format
|
||||||
|
|
||||||
|
### Supported Export Formats
|
||||||
|
|
||||||
|
- **TXT** - Plain text (content only)
|
||||||
|
- **SRT** - SubRip subtitle format (with timestamps)
|
||||||
|
- **VTT** - WebVTT subtitle format (with timestamps)
|
||||||
|
- **JSON** - Structured format with segments and metadata
|
||||||
|
- **TSV** - Tab-separated values (spreadsheet compatible)
|
||||||
|
|
||||||
|
## Configuration
|
||||||
|
|
||||||
|
Edit `config.py` to customize:
|
||||||
|
|
||||||
|
```python
|
||||||
|
# Model size (tiny, base, small, medium, large)
|
||||||
|
DEFAULT_MODEL = "medium"
|
||||||
|
|
||||||
|
# Language code
|
||||||
|
LANGUAGE_CODE = "fa" # Farsi
|
||||||
|
|
||||||
|
# Supported formats
|
||||||
|
SUPPORTED_AUDIO_FORMATS = {".mp3", ".wav", ".m4a", ".flac", ".ogg", ...}
|
||||||
|
SUPPORTED_VIDEO_FORMATS = {".mp4", ".mkv", ".mov", ".webm", ".avi", ...}
|
||||||
|
```
|
||||||
|
|
||||||
|
## Model Information
|
||||||
|
|
||||||
|
### Available Models
|
||||||
|
|
||||||
|
| Model | Size | Speed | Accuracy | VRAM |
|
||||||
|
|-------|------|-------|----------|------|
|
||||||
|
| tiny | 39M | ~10x | Good | ~1GB |
|
||||||
|
| base | 74M | ~7x | Very Good | ~1GB |
|
||||||
|
| small | 244M | ~4x | Excellent | ~2GB |
|
||||||
|
| medium | 769M | ~2x | Excellent | ~5GB |
|
||||||
|
| large | 1550M | 1x | Best | ~10GB |
|
||||||
|
|
||||||
|
**Default**: `medium` (recommended for Farsi)
|
||||||
|
|
||||||
|
### Performance Notes
|
||||||
|
|
||||||
|
- Larger models provide better accuracy but require more VRAM
|
||||||
|
- GPU (CUDA) dramatically speeds up transcription (8-10x faster)
|
||||||
|
- First run downloads the model (~500MB-3GB depending on model size)
|
||||||
|
- Subsequent runs use cached model files
|
||||||
|
|
||||||
|
## Project Structure
|
||||||
|
|
||||||
|
```
|
||||||
|
farsi_transcriber/
|
||||||
|
├── ui/ # User interface components
|
||||||
|
│ ├── __init__.py
|
||||||
|
│ ├── main_window.py # Main application window
|
||||||
|
│ └── styles.py # Styling and theming
|
||||||
|
├── models/ # Model management
|
||||||
|
│ ├── __init__.py
|
||||||
|
│ └── whisper_transcriber.py # Whisper wrapper
|
||||||
|
├── utils/ # Utility functions
|
||||||
|
│ ├── __init__.py
|
||||||
|
│ └── export.py # Export functionality
|
||||||
|
├── config.py # Configuration settings
|
||||||
|
├── main.py # Application entry point
|
||||||
|
├── __init__.py # Package init
|
||||||
|
├── requirements.txt # Python dependencies
|
||||||
|
└── README.md # This file
|
||||||
|
```
|
||||||
|
|
||||||
|
## Troubleshooting
|
||||||
|
|
||||||
|
### Issue: "ffmpeg not found"
|
||||||
|
**Solution**: Install ffmpeg using your package manager (see Installation section)
|
||||||
|
|
||||||
|
### Issue: "CUDA out of memory"
|
||||||
|
**Solution**: Use a smaller model or reduce audio processing in chunks
|
||||||
|
|
||||||
|
### Issue: "Model download fails"
|
||||||
|
**Solution**: Check internet connection, try again. Models are cached in `~/.cache/whisper/`
|
||||||
|
|
||||||
|
### Issue: Slow transcription
|
||||||
|
**Solution**: Ensure CUDA is detected (`nvidia-smi`), or upgrade to a smaller/faster model
|
||||||
|
|
||||||
|
## Advanced Usage
|
||||||
|
|
||||||
|
### Custom Model Selection
|
||||||
|
|
||||||
|
Update `config.py`:
|
||||||
|
```python
|
||||||
|
DEFAULT_MODEL = "large" # For maximum accuracy
|
||||||
|
# or
|
||||||
|
DEFAULT_MODEL = "tiny" # For fastest processing
|
||||||
|
```
|
||||||
|
|
||||||
|
### Batch Processing (Future)
|
||||||
|
|
||||||
|
Script to process multiple files:
|
||||||
|
```python
|
||||||
|
from farsi_transcriber.models.whisper_transcriber import FarsiTranscriber
|
||||||
|
|
||||||
|
transcriber = FarsiTranscriber(model_name="medium")
|
||||||
|
for audio_file in audio_files:
|
||||||
|
result = transcriber.transcribe(audio_file)
|
||||||
|
# Process results
|
||||||
|
```
|
||||||
|
|
||||||
|
## Performance Tips
|
||||||
|
|
||||||
|
1. **Use GPU** - Ensure NVIDIA CUDA is properly installed
|
||||||
|
2. **Choose appropriate model** - Balance speed vs accuracy
|
||||||
|
3. **Close other applications** - Free up RAM/VRAM
|
||||||
|
4. **Use SSD** - Faster model loading and temporary file I/O
|
||||||
|
5. **Local processing** - All processing happens locally, no cloud uploads
|
||||||
|
|
||||||
|
## Development
|
||||||
|
|
||||||
|
### Code Style
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# Format code
|
||||||
|
black farsi_transcriber/
|
||||||
|
|
||||||
|
# Check style
|
||||||
|
flake8 farsi_transcriber/
|
||||||
|
|
||||||
|
# Sort imports
|
||||||
|
isort farsi_transcriber/
|
||||||
|
```
|
||||||
|
|
||||||
|
### Future Features
|
||||||
|
|
||||||
|
- [ ] Batch processing
|
||||||
|
- [ ] Real-time transcription preview
|
||||||
|
- [ ] Speaker diarization
|
||||||
|
- [ ] Multi-language support UI
|
||||||
|
- [ ] Settings dialog
|
||||||
|
- [ ] Keyboard shortcuts
|
||||||
|
- [ ] Drag-and-drop support
|
||||||
|
- [ ] Recent files history
|
||||||
|
|
||||||
|
## License
|
||||||
|
|
||||||
|
MIT License - Personal use and modifications allowed
|
||||||
|
|
||||||
|
## Acknowledgments
|
||||||
|
|
||||||
|
Built with:
|
||||||
|
- [OpenAI Whisper](https://github.com/openai/whisper) - Speech recognition
|
||||||
|
- [PyQt6](https://www.riverbankcomputing.com/software/pyqt/) - GUI framework
|
||||||
|
- [PyTorch](https://pytorch.org/) - Deep learning
|
||||||
|
|
||||||
|
## Support
|
||||||
|
|
||||||
|
For issues or suggestions:
|
||||||
|
1. Check the troubleshooting section
|
||||||
|
2. Verify ffmpeg is installed
|
||||||
|
3. Ensure Python 3.8+ is used
|
||||||
|
4. Check available disk space
|
||||||
|
5. Verify CUDA setup (for GPU users)
|
||||||
8
farsi_transcriber/__init__.py
Normal file
8
farsi_transcriber/__init__.py
Normal file
@ -0,0 +1,8 @@
|
|||||||
|
"""
|
||||||
|
Farsi Transcriber Application
|
||||||
|
|
||||||
|
A desktop application for transcribing Farsi audio and video files using OpenAI's Whisper.
|
||||||
|
"""
|
||||||
|
|
||||||
|
__version__ = "0.1.0"
|
||||||
|
__author__ = "Personal Project"
|
||||||
71
farsi_transcriber/config.py
Normal file
71
farsi_transcriber/config.py
Normal file
@ -0,0 +1,71 @@
|
|||||||
|
"""
|
||||||
|
Configuration settings for Farsi Transcriber application
|
||||||
|
|
||||||
|
Manages model selection, device settings, and other configuration options.
|
||||||
|
"""
|
||||||
|
|
||||||
|
from pathlib import Path
|
||||||
|
|
||||||
|
# Application metadata
|
||||||
|
APP_NAME = "Farsi Transcriber"
|
||||||
|
APP_VERSION = "0.1.0"
|
||||||
|
APP_DESCRIPTION = "A desktop application for transcribing Farsi audio and video files"
|
||||||
|
|
||||||
|
# Model settings
|
||||||
|
DEFAULT_MODEL = "medium" # Options: tiny, base, small, medium, large
|
||||||
|
AVAILABLE_MODELS = ["tiny", "base", "small", "medium", "large"]
|
||||||
|
MODEL_DESCRIPTIONS = {
|
||||||
|
"tiny": "Smallest model (39M params) - Fastest, ~1GB VRAM required",
|
||||||
|
"base": "Small model (74M params) - Fast, ~1GB VRAM required",
|
||||||
|
"small": "Medium model (244M params) - Balanced, ~2GB VRAM required",
|
||||||
|
"medium": "Large model (769M params) - Good accuracy, ~5GB VRAM required",
|
||||||
|
"large": "Largest model (1550M params) - Best accuracy, ~10GB VRAM required",
|
||||||
|
}
|
||||||
|
|
||||||
|
# Language settings
|
||||||
|
LANGUAGE_CODE = "fa" # Farsi/Persian
|
||||||
|
LANGUAGE_NAME = "Farsi"
|
||||||
|
|
||||||
|
# Audio/Video settings
|
||||||
|
SUPPORTED_AUDIO_FORMATS = {".mp3", ".wav", ".m4a", ".flac", ".ogg", ".aac", ".wma"}
|
||||||
|
SUPPORTED_VIDEO_FORMATS = {".mp4", ".mkv", ".mov", ".webm", ".avi", ".flv", ".wmv"}
|
||||||
|
|
||||||
|
# UI settings
|
||||||
|
WINDOW_WIDTH = 900
|
||||||
|
WINDOW_HEIGHT = 700
|
||||||
|
WINDOW_MIN_WIDTH = 800
|
||||||
|
WINDOW_MIN_HEIGHT = 600
|
||||||
|
|
||||||
|
# Output settings
|
||||||
|
OUTPUT_DIR = Path.home() / "FarsiTranscriber" / "outputs"
|
||||||
|
OUTPUT_DIR.mkdir(parents=True, exist_ok=True)
|
||||||
|
|
||||||
|
EXPORT_FORMATS = {
|
||||||
|
"txt": "Plain Text",
|
||||||
|
"srt": "SRT Subtitles",
|
||||||
|
"vtt": "WebVTT Subtitles",
|
||||||
|
"json": "JSON Format",
|
||||||
|
"tsv": "Tab-Separated Values",
|
||||||
|
}
|
||||||
|
|
||||||
|
# Device settings (auto-detect CUDA if available)
|
||||||
|
try:
|
||||||
|
import torch
|
||||||
|
|
||||||
|
DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
|
||||||
|
except ImportError:
|
||||||
|
DEVICE = "cpu"
|
||||||
|
|
||||||
|
# Logging settings
|
||||||
|
LOG_LEVEL = "INFO"
|
||||||
|
LOG_FILE = OUTPUT_DIR / "transcriber.log"
|
||||||
|
|
||||||
|
|
||||||
|
def get_model_info(model_name: str) -> str:
|
||||||
|
"""Get description for a model"""
|
||||||
|
return MODEL_DESCRIPTIONS.get(model_name, "Unknown model")
|
||||||
|
|
||||||
|
|
||||||
|
def get_supported_formats() -> set:
|
||||||
|
"""Get all supported audio and video formats"""
|
||||||
|
return SUPPORTED_AUDIO_FORMATS | SUPPORTED_VIDEO_FORMATS
|
||||||
26
farsi_transcriber/main.py
Normal file
26
farsi_transcriber/main.py
Normal file
@ -0,0 +1,26 @@
|
|||||||
|
#!/usr/bin/env python3
|
||||||
|
"""
|
||||||
|
Farsi Transcriber - Main Entry Point
|
||||||
|
|
||||||
|
A PyQt6-based desktop application for transcribing Farsi audio and video files.
|
||||||
|
"""
|
||||||
|
|
||||||
|
import sys
|
||||||
|
from PyQt6.QtWidgets import QApplication
|
||||||
|
|
||||||
|
from farsi_transcriber.ui.main_window import MainWindow
|
||||||
|
|
||||||
|
|
||||||
|
def main():
|
||||||
|
"""Main entry point for the application"""
|
||||||
|
app = QApplication(sys.argv)
|
||||||
|
|
||||||
|
# Create and show main window
|
||||||
|
window = MainWindow()
|
||||||
|
window.show()
|
||||||
|
|
||||||
|
sys.exit(app.exec())
|
||||||
|
|
||||||
|
|
||||||
|
if __name__ == "__main__":
|
||||||
|
main()
|
||||||
1
farsi_transcriber/models/__init__.py
Normal file
1
farsi_transcriber/models/__init__.py
Normal file
@ -0,0 +1 @@
|
|||||||
|
"""Model management for Farsi Transcriber"""
|
||||||
225
farsi_transcriber/models/whisper_transcriber.py
Normal file
225
farsi_transcriber/models/whisper_transcriber.py
Normal file
@ -0,0 +1,225 @@
|
|||||||
|
"""
|
||||||
|
Whisper Transcriber Module
|
||||||
|
|
||||||
|
Handles Farsi audio/video transcription using OpenAI's Whisper model.
|
||||||
|
"""
|
||||||
|
|
||||||
|
import warnings
|
||||||
|
from pathlib import Path
|
||||||
|
from typing import Dict, List, Optional
|
||||||
|
|
||||||
|
import torch
|
||||||
|
import whisper
|
||||||
|
|
||||||
|
|
||||||
|
class FarsiTranscriber:
|
||||||
|
"""
|
||||||
|
Wrapper around Whisper model for Farsi transcription.
|
||||||
|
|
||||||
|
Supports both audio and video files, with word-level timestamp extraction.
|
||||||
|
"""
|
||||||
|
|
||||||
|
# Supported audio formats
|
||||||
|
AUDIO_FORMATS = {".mp3", ".wav", ".m4a", ".flac", ".ogg", ".aac", ".wma"}
|
||||||
|
|
||||||
|
# Supported video formats
|
||||||
|
VIDEO_FORMATS = {".mp4", ".mkv", ".mov", ".webm", ".avi", ".flv", ".wmv"}
|
||||||
|
|
||||||
|
# Language code for Farsi/Persian
|
||||||
|
FARSI_LANGUAGE = "fa"
|
||||||
|
|
||||||
|
def __init__(self, model_name: str = "medium", device: Optional[str] = None):
|
||||||
|
"""
|
||||||
|
Initialize Farsi Transcriber.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
model_name: Whisper model size ('tiny', 'base', 'small', 'medium', 'large')
|
||||||
|
device: Device to use ('cuda', 'cpu'). Auto-detect if None.
|
||||||
|
"""
|
||||||
|
self.model_name = model_name
|
||||||
|
|
||||||
|
# Auto-detect device
|
||||||
|
if device is None:
|
||||||
|
self.device = "cuda" if torch.cuda.is_available() else "cpu"
|
||||||
|
else:
|
||||||
|
self.device = device
|
||||||
|
|
||||||
|
print(f"Using device: {self.device}")
|
||||||
|
|
||||||
|
# Load model
|
||||||
|
print(f"Loading Whisper model: {model_name}...")
|
||||||
|
with warnings.catch_warnings():
|
||||||
|
warnings.simplefilter("ignore")
|
||||||
|
self.model = whisper.load_model(model_name, device=self.device)
|
||||||
|
|
||||||
|
print("Model loaded successfully")
|
||||||
|
|
||||||
|
def transcribe(
|
||||||
|
self,
|
||||||
|
file_path: str,
|
||||||
|
language: str = FARSI_LANGUAGE,
|
||||||
|
verbose: bool = False,
|
||||||
|
) -> Dict:
|
||||||
|
"""
|
||||||
|
Transcribe an audio or video file in Farsi.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
file_path: Path to audio or video file
|
||||||
|
language: Language code (default: 'fa' for Farsi)
|
||||||
|
verbose: Whether to print progress
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
Dictionary with transcription results including word-level segments
|
||||||
|
"""
|
||||||
|
file_path = Path(file_path)
|
||||||
|
|
||||||
|
# Validate file exists
|
||||||
|
if not file_path.exists():
|
||||||
|
raise FileNotFoundError(f"File not found: {file_path}")
|
||||||
|
|
||||||
|
# Check format is supported
|
||||||
|
if not self._is_supported_format(file_path):
|
||||||
|
raise ValueError(
|
||||||
|
f"Unsupported format: {file_path.suffix}. "
|
||||||
|
f"Supported: {self.AUDIO_FORMATS | self.VIDEO_FORMATS}"
|
||||||
|
)
|
||||||
|
|
||||||
|
# Perform transcription
|
||||||
|
print(f"Transcribing: {file_path.name}")
|
||||||
|
|
||||||
|
result = self.model.transcribe(
|
||||||
|
str(file_path),
|
||||||
|
language=language,
|
||||||
|
verbose=verbose,
|
||||||
|
)
|
||||||
|
|
||||||
|
# Enhance result with word-level segments
|
||||||
|
enhanced_result = self._enhance_with_word_segments(result)
|
||||||
|
|
||||||
|
return enhanced_result
|
||||||
|
|
||||||
|
def _is_supported_format(self, file_path: Path) -> bool:
|
||||||
|
"""Check if file format is supported."""
|
||||||
|
suffix = file_path.suffix.lower()
|
||||||
|
return suffix in (self.AUDIO_FORMATS | self.VIDEO_FORMATS)
|
||||||
|
|
||||||
|
def _enhance_with_word_segments(self, result: Dict) -> Dict:
|
||||||
|
"""
|
||||||
|
Enhance transcription result with word-level timing information.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Whisper transcription result
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
Enhanced result with word-level segments
|
||||||
|
"""
|
||||||
|
enhanced_segments = []
|
||||||
|
|
||||||
|
for segment in result.get("segments", []):
|
||||||
|
# Extract word-level timing if available
|
||||||
|
word_segments = self._extract_word_segments(segment)
|
||||||
|
|
||||||
|
enhanced_segment = {
|
||||||
|
"id": segment.get("id"),
|
||||||
|
"start": segment.get("start"),
|
||||||
|
"end": segment.get("end"),
|
||||||
|
"text": segment.get("text", ""),
|
||||||
|
"words": word_segments,
|
||||||
|
}
|
||||||
|
enhanced_segments.append(enhanced_segment)
|
||||||
|
|
||||||
|
result["segments"] = enhanced_segments
|
||||||
|
return result
|
||||||
|
|
||||||
|
def _extract_word_segments(self, segment: Dict) -> List[Dict]:
|
||||||
|
"""
|
||||||
|
Extract word-level timing from a segment.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
segment: Whisper segment with text
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
List of word dictionaries with timing information
|
||||||
|
"""
|
||||||
|
text = segment.get("text", "").strip()
|
||||||
|
if not text:
|
||||||
|
return []
|
||||||
|
|
||||||
|
# For now, return simple word list
|
||||||
|
# Whisper v3 includes word-level details in some configurations
|
||||||
|
start_time = segment.get("start", 0)
|
||||||
|
end_time = segment.get("end", 0)
|
||||||
|
duration = end_time - start_time
|
||||||
|
|
||||||
|
words = text.split()
|
||||||
|
if not words:
|
||||||
|
return []
|
||||||
|
|
||||||
|
# Distribute time evenly across words (simple approach)
|
||||||
|
# More sophisticated timing can be extracted from Whisper's internal data
|
||||||
|
word_duration = duration / len(words) if words else 0
|
||||||
|
|
||||||
|
word_segments = []
|
||||||
|
for i, word in enumerate(words):
|
||||||
|
word_start = start_time + (i * word_duration)
|
||||||
|
word_end = word_start + word_duration
|
||||||
|
|
||||||
|
word_segments.append(
|
||||||
|
{
|
||||||
|
"word": word,
|
||||||
|
"start": word_start,
|
||||||
|
"end": word_end,
|
||||||
|
}
|
||||||
|
)
|
||||||
|
|
||||||
|
return word_segments
|
||||||
|
|
||||||
|
def format_result_for_display(
|
||||||
|
self, result: Dict, include_timestamps: bool = True
|
||||||
|
) -> str:
|
||||||
|
"""
|
||||||
|
Format transcription result for display in UI.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Transcription result
|
||||||
|
include_timestamps: Whether to include timestamps
|
||||||
|
|
||||||
|
Returns:
|
||||||
|
Formatted text string
|
||||||
|
"""
|
||||||
|
lines = []
|
||||||
|
|
||||||
|
for segment in result.get("segments", []):
|
||||||
|
text = segment.get("text", "").strip()
|
||||||
|
if not text:
|
||||||
|
continue
|
||||||
|
|
||||||
|
if include_timestamps:
|
||||||
|
start = segment.get("start", 0)
|
||||||
|
end = segment.get("end", 0)
|
||||||
|
timestamp = f"[{self._format_time(start)} - {self._format_time(end)}]"
|
||||||
|
lines.append(f"{timestamp}\n{text}\n")
|
||||||
|
else:
|
||||||
|
lines.append(text)
|
||||||
|
|
||||||
|
return "\n".join(lines)
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def _format_time(seconds: float) -> str:
|
||||||
|
"""Format seconds to HH:MM:SS format."""
|
||||||
|
hours = int(seconds // 3600)
|
||||||
|
minutes = int((seconds % 3600) // 60)
|
||||||
|
secs = int(seconds % 60)
|
||||||
|
milliseconds = int((seconds % 1) * 1000)
|
||||||
|
|
||||||
|
return f"{hours:02d}:{minutes:02d}:{secs:02d}.{milliseconds:03d}"
|
||||||
|
|
||||||
|
def get_device_info(self) -> str:
|
||||||
|
"""Get information about current device and model."""
|
||||||
|
return (
|
||||||
|
f"Model: {self.model_name} | "
|
||||||
|
f"Device: {self.device.upper()} | "
|
||||||
|
f"VRAM: {torch.cuda.get_device_properties(self.device).total_memory / 1e9:.1f}GB "
|
||||||
|
if self.device == "cuda"
|
||||||
|
else f"Model: {self.model_name} | Device: {self.device.upper()}"
|
||||||
|
)
|
||||||
7
farsi_transcriber/requirements.txt
Normal file
7
farsi_transcriber/requirements.txt
Normal file
@ -0,0 +1,7 @@
|
|||||||
|
PyQt6==6.6.1
|
||||||
|
PyQt6-Qt6==6.6.1
|
||||||
|
PyQt6-sip==13.6.0
|
||||||
|
torch>=1.10.1
|
||||||
|
numpy
|
||||||
|
openai-whisper
|
||||||
|
tqdm
|
||||||
1
farsi_transcriber/ui/__init__.py
Normal file
1
farsi_transcriber/ui/__init__.py
Normal file
@ -0,0 +1 @@
|
|||||||
|
"""UI components for Farsi Transcriber"""
|
||||||
285
farsi_transcriber/ui/main_window.py
Normal file
285
farsi_transcriber/ui/main_window.py
Normal file
@ -0,0 +1,285 @@
|
|||||||
|
"""
|
||||||
|
Main application window for Farsi Transcriber
|
||||||
|
|
||||||
|
Provides PyQt6-based GUI for selecting files and transcribing Farsi audio/video.
|
||||||
|
"""
|
||||||
|
|
||||||
|
import os
|
||||||
|
from pathlib import Path
|
||||||
|
|
||||||
|
from PyQt6.QtCore import Qt, QThread, pyqtSignal
|
||||||
|
from PyQt6.QtWidgets import (
|
||||||
|
QMainWindow,
|
||||||
|
QWidget,
|
||||||
|
QVBoxLayout,
|
||||||
|
QHBoxLayout,
|
||||||
|
QPushButton,
|
||||||
|
QLabel,
|
||||||
|
QTextEdit,
|
||||||
|
QProgressBar,
|
||||||
|
QFileDialog,
|
||||||
|
QMessageBox,
|
||||||
|
)
|
||||||
|
from PyQt6.QtGui import QFont
|
||||||
|
|
||||||
|
from farsi_transcriber.models.whisper_transcriber import FarsiTranscriber
|
||||||
|
from farsi_transcriber.utils.export import TranscriptionExporter
|
||||||
|
from farsi_transcriber.ui.styles import get_stylesheet, get_color
|
||||||
|
|
||||||
|
|
||||||
|
class TranscriptionWorker(QThread):
|
||||||
|
"""Worker thread for transcription to prevent UI freezing"""
|
||||||
|
|
||||||
|
# Signals
|
||||||
|
progress_update = pyqtSignal(str) # Status messages
|
||||||
|
transcription_complete = pyqtSignal(dict) # Results with timestamps
|
||||||
|
error_occurred = pyqtSignal(str) # Error messages
|
||||||
|
|
||||||
|
def __init__(self, file_path: str, model_name: str = "medium"):
|
||||||
|
super().__init__()
|
||||||
|
self.file_path = file_path
|
||||||
|
self.model_name = model_name
|
||||||
|
self.transcriber = None
|
||||||
|
|
||||||
|
def run(self):
|
||||||
|
"""Run transcription in background thread"""
|
||||||
|
try:
|
||||||
|
# Initialize Whisper transcriber
|
||||||
|
self.progress_update.emit("Loading Whisper model...")
|
||||||
|
self.transcriber = FarsiTranscriber(model_name=self.model_name)
|
||||||
|
|
||||||
|
# Perform transcription
|
||||||
|
self.progress_update.emit(f"Transcribing: {Path(self.file_path).name}")
|
||||||
|
result = self.transcriber.transcribe(self.file_path)
|
||||||
|
|
||||||
|
# Format result for display with timestamps
|
||||||
|
display_text = self.transcriber.format_result_for_display(result)
|
||||||
|
|
||||||
|
# Add full text for export
|
||||||
|
result["full_text"] = result.get("text", "")
|
||||||
|
|
||||||
|
self.progress_update.emit("Transcription complete!")
|
||||||
|
self.transcription_complete.emit(
|
||||||
|
{
|
||||||
|
"text": display_text,
|
||||||
|
"segments": result.get("segments", []),
|
||||||
|
"full_text": result.get("text", ""),
|
||||||
|
}
|
||||||
|
)
|
||||||
|
|
||||||
|
except Exception as e:
|
||||||
|
self.error_occurred.emit(f"Error: {str(e)}")
|
||||||
|
|
||||||
|
|
||||||
|
class MainWindow(QMainWindow):
|
||||||
|
"""Main application window for Farsi Transcriber"""
|
||||||
|
|
||||||
|
# Supported audio and video formats
|
||||||
|
SUPPORTED_FORMATS = (
|
||||||
|
"Audio Files (*.mp3 *.wav *.m4a *.flac *.ogg *.aac *.wma);;",
|
||||||
|
"Video Files (*.mp4 *.mkv *.mov *.webm *.avi *.flv *.wmv);;",
|
||||||
|
"All Files (*.*)",
|
||||||
|
)
|
||||||
|
|
||||||
|
def __init__(self):
|
||||||
|
super().__init__()
|
||||||
|
self.selected_file = None
|
||||||
|
self.transcription_worker = None
|
||||||
|
self.last_result = None
|
||||||
|
# Apply stylesheet
|
||||||
|
self.setStyleSheet(get_stylesheet())
|
||||||
|
self.init_ui()
|
||||||
|
|
||||||
|
def init_ui(self):
|
||||||
|
"""Initialize the user interface"""
|
||||||
|
self.setWindowTitle("Farsi Transcriber")
|
||||||
|
self.setGeometry(100, 100, 900, 700)
|
||||||
|
|
||||||
|
# Create central widget and main layout
|
||||||
|
central_widget = QWidget()
|
||||||
|
self.setCentralWidget(central_widget)
|
||||||
|
main_layout = QVBoxLayout(central_widget)
|
||||||
|
main_layout.setSpacing(10)
|
||||||
|
main_layout.setContentsMargins(20, 20, 20, 20)
|
||||||
|
|
||||||
|
# Title
|
||||||
|
title_label = QLabel("Farsi Audio/Video Transcriber")
|
||||||
|
title_font = QFont()
|
||||||
|
title_font.setPointSize(16)
|
||||||
|
title_font.setBold(True)
|
||||||
|
title_label.setFont(title_font)
|
||||||
|
main_layout.addWidget(title_label)
|
||||||
|
|
||||||
|
# File selection section
|
||||||
|
file_section_layout = QHBoxLayout()
|
||||||
|
self.file_label = QLabel("No file selected")
|
||||||
|
self.file_label.setStyleSheet("color: gray;")
|
||||||
|
file_section_layout.addWidget(self.file_label, 1)
|
||||||
|
|
||||||
|
self.select_button = QPushButton("Select File")
|
||||||
|
self.select_button.clicked.connect(self.on_select_file)
|
||||||
|
file_section_layout.addWidget(self.select_button)
|
||||||
|
|
||||||
|
main_layout.addLayout(file_section_layout)
|
||||||
|
|
||||||
|
# Transcribe button
|
||||||
|
self.transcribe_button = QPushButton("Transcribe")
|
||||||
|
self.transcribe_button.clicked.connect(self.on_transcribe)
|
||||||
|
self.transcribe_button.setEnabled(False)
|
||||||
|
main_layout.addWidget(self.transcribe_button)
|
||||||
|
|
||||||
|
# Progress bar
|
||||||
|
self.progress_bar = QProgressBar()
|
||||||
|
self.progress_bar.setRange(0, 0) # Indeterminate progress
|
||||||
|
self.progress_bar.setVisible(False)
|
||||||
|
main_layout.addWidget(self.progress_bar)
|
||||||
|
|
||||||
|
# Status label
|
||||||
|
self.status_label = QLabel("Ready")
|
||||||
|
self.status_label.setStyleSheet("color: #666; font-style: italic;")
|
||||||
|
main_layout.addWidget(self.status_label)
|
||||||
|
|
||||||
|
# Results text area
|
||||||
|
results_title = QLabel("Transcription Results:")
|
||||||
|
results_title_font = QFont()
|
||||||
|
results_title_font.setBold(True)
|
||||||
|
results_title.setFont(results_title_font)
|
||||||
|
main_layout.addWidget(results_title)
|
||||||
|
|
||||||
|
self.results_text = QTextEdit()
|
||||||
|
self.results_text.setReadOnly(True)
|
||||||
|
self.results_text.setPlaceholderText(
|
||||||
|
"Transcription results will appear here..."
|
||||||
|
)
|
||||||
|
# Set monospace font for results
|
||||||
|
mono_font = QFont("Courier New", 10)
|
||||||
|
self.results_text.setFont(mono_font)
|
||||||
|
main_layout.addWidget(self.results_text)
|
||||||
|
|
||||||
|
# Buttons layout (Export, Clear)
|
||||||
|
buttons_layout = QHBoxLayout()
|
||||||
|
buttons_layout.addStretch()
|
||||||
|
|
||||||
|
self.export_button = QPushButton("Export Results")
|
||||||
|
self.export_button.clicked.connect(self.on_export)
|
||||||
|
self.export_button.setEnabled(False)
|
||||||
|
buttons_layout.addWidget(self.export_button)
|
||||||
|
|
||||||
|
self.clear_button = QPushButton("Clear")
|
||||||
|
self.clear_button.clicked.connect(self.on_clear)
|
||||||
|
buttons_layout.addWidget(self.clear_button)
|
||||||
|
|
||||||
|
main_layout.addLayout(buttons_layout)
|
||||||
|
|
||||||
|
def on_select_file(self):
|
||||||
|
"""Handle file selection dialog"""
|
||||||
|
file_path, _ = QFileDialog.getOpenFileName(
|
||||||
|
self, "Select Audio or Video File", "", "".join(self.SUPPORTED_FORMATS)
|
||||||
|
)
|
||||||
|
|
||||||
|
if file_path:
|
||||||
|
self.selected_file = file_path
|
||||||
|
file_name = Path(file_path).name
|
||||||
|
self.file_label.setText(f"Selected: {file_name}")
|
||||||
|
self.file_label.setStyleSheet("color: #333;")
|
||||||
|
self.transcribe_button.setEnabled(True)
|
||||||
|
self.export_button.setEnabled(False)
|
||||||
|
self.results_text.clear()
|
||||||
|
self.status_label.setText("File selected. Click 'Transcribe' to start.")
|
||||||
|
|
||||||
|
def on_transcribe(self):
|
||||||
|
"""Handle transcription button click"""
|
||||||
|
if not self.selected_file:
|
||||||
|
QMessageBox.warning(self, "Error", "Please select a file first.")
|
||||||
|
return
|
||||||
|
|
||||||
|
# Disable buttons during transcription
|
||||||
|
self.transcribe_button.setEnabled(False)
|
||||||
|
self.select_button.setEnabled(False)
|
||||||
|
self.export_button.setEnabled(False)
|
||||||
|
|
||||||
|
# Show progress
|
||||||
|
self.progress_bar.setVisible(True)
|
||||||
|
self.status_label.setText("Transcribing...")
|
||||||
|
|
||||||
|
# Create and start worker thread
|
||||||
|
self.transcription_worker = TranscriptionWorker(self.selected_file)
|
||||||
|
self.transcription_worker.progress_update.connect(self.on_progress_update)
|
||||||
|
self.transcription_worker.transcription_complete.connect(
|
||||||
|
self.on_transcription_complete
|
||||||
|
)
|
||||||
|
self.transcription_worker.error_occurred.connect(self.on_error)
|
||||||
|
self.transcription_worker.start()
|
||||||
|
|
||||||
|
def on_progress_update(self, message: str):
|
||||||
|
"""Handle progress updates from worker thread"""
|
||||||
|
self.status_label.setText(message)
|
||||||
|
|
||||||
|
def on_transcription_complete(self, result: dict):
|
||||||
|
"""Handle completed transcription"""
|
||||||
|
self.progress_bar.setVisible(False)
|
||||||
|
self.transcribe_button.setEnabled(True)
|
||||||
|
self.select_button.setEnabled(True)
|
||||||
|
self.export_button.setEnabled(True)
|
||||||
|
self.status_label.setText("Transcription complete!")
|
||||||
|
|
||||||
|
# Display results with timestamps
|
||||||
|
self.results_text.setText(result.get("text", "No transcription available"))
|
||||||
|
|
||||||
|
# Store result for export
|
||||||
|
self.last_result = result
|
||||||
|
|
||||||
|
def on_error(self, error_message: str):
|
||||||
|
"""Handle errors from worker thread"""
|
||||||
|
self.progress_bar.setVisible(False)
|
||||||
|
self.transcribe_button.setEnabled(True)
|
||||||
|
self.select_button.setEnabled(True)
|
||||||
|
self.status_label.setText("Error occurred. Check message below.")
|
||||||
|
QMessageBox.critical(self, "Transcription Error", error_message)
|
||||||
|
|
||||||
|
def on_export(self):
|
||||||
|
"""Handle export button click"""
|
||||||
|
if not self.last_result:
|
||||||
|
QMessageBox.warning(self, "Warning", "No transcription to export.")
|
||||||
|
return
|
||||||
|
|
||||||
|
file_path, file_filter = QFileDialog.getSaveFileName(
|
||||||
|
self,
|
||||||
|
"Export Transcription",
|
||||||
|
"",
|
||||||
|
"Text Files (*.txt);;SRT Subtitles (*.srt);;WebVTT Subtitles (*.vtt);;JSON (*.json);;TSV (*.tsv)",
|
||||||
|
)
|
||||||
|
|
||||||
|
if file_path:
|
||||||
|
try:
|
||||||
|
file_path = Path(file_path)
|
||||||
|
|
||||||
|
# Determine format from file extension
|
||||||
|
suffix = file_path.suffix.lower().lstrip(".")
|
||||||
|
if not suffix:
|
||||||
|
# Default to txt if no extension
|
||||||
|
suffix = "txt"
|
||||||
|
file_path = file_path.with_suffix(".txt")
|
||||||
|
|
||||||
|
# Export using the appropriate format
|
||||||
|
TranscriptionExporter.export(self.last_result, file_path, suffix)
|
||||||
|
|
||||||
|
QMessageBox.information(
|
||||||
|
self,
|
||||||
|
"Success",
|
||||||
|
f"Transcription exported successfully to:\n{file_path.name}",
|
||||||
|
)
|
||||||
|
except Exception as e:
|
||||||
|
QMessageBox.critical(
|
||||||
|
self, "Export Error", f"Failed to export: {str(e)}"
|
||||||
|
)
|
||||||
|
|
||||||
|
def on_clear(self):
|
||||||
|
"""Clear all results and reset UI"""
|
||||||
|
self.selected_file = None
|
||||||
|
self.file_label.setText("No file selected")
|
||||||
|
self.file_label.setStyleSheet("color: gray;")
|
||||||
|
self.results_text.clear()
|
||||||
|
self.status_label.setText("Ready")
|
||||||
|
self.transcribe_button.setEnabled(False)
|
||||||
|
self.export_button.setEnabled(False)
|
||||||
107
farsi_transcriber/ui/styles.py
Normal file
107
farsi_transcriber/ui/styles.py
Normal file
@ -0,0 +1,107 @@
|
|||||||
|
"""
|
||||||
|
Application styling and theming
|
||||||
|
|
||||||
|
Provides stylesheet and styling utilities for the Farsi Transcriber app.
|
||||||
|
"""
|
||||||
|
|
||||||
|
# Modern, professional dark-themed stylesheet
|
||||||
|
MAIN_STYLESHEET = """
|
||||||
|
QMainWindow {
|
||||||
|
background-color: #f5f5f5;
|
||||||
|
}
|
||||||
|
|
||||||
|
QLabel {
|
||||||
|
color: #333333;
|
||||||
|
}
|
||||||
|
|
||||||
|
QLineEdit, QTextEdit {
|
||||||
|
background-color: #ffffff;
|
||||||
|
color: #333333;
|
||||||
|
border: 1px solid #d0d0d0;
|
||||||
|
border-radius: 4px;
|
||||||
|
padding: 5px;
|
||||||
|
font-size: 11pt;
|
||||||
|
}
|
||||||
|
|
||||||
|
QLineEdit:focus, QTextEdit:focus {
|
||||||
|
border: 2px solid #4CAF50;
|
||||||
|
background-color: #fafafa;
|
||||||
|
}
|
||||||
|
|
||||||
|
QPushButton {
|
||||||
|
background-color: #4CAF50;
|
||||||
|
color: white;
|
||||||
|
border: none;
|
||||||
|
border-radius: 4px;
|
||||||
|
padding: 8px 16px;
|
||||||
|
font-weight: bold;
|
||||||
|
font-size: 11pt;
|
||||||
|
min-height: 32px;
|
||||||
|
}
|
||||||
|
|
||||||
|
QPushButton:hover {
|
||||||
|
background-color: #45a049;
|
||||||
|
}
|
||||||
|
|
||||||
|
QPushButton:pressed {
|
||||||
|
background-color: #3d8b40;
|
||||||
|
}
|
||||||
|
|
||||||
|
QPushButton:disabled {
|
||||||
|
background-color: #cccccc;
|
||||||
|
color: #999999;
|
||||||
|
}
|
||||||
|
|
||||||
|
QProgressBar {
|
||||||
|
border: 1px solid #d0d0d0;
|
||||||
|
border-radius: 4px;
|
||||||
|
text-align: center;
|
||||||
|
background-color: #ffffff;
|
||||||
|
height: 20px;
|
||||||
|
}
|
||||||
|
|
||||||
|
QProgressBar::chunk {
|
||||||
|
background-color: #4CAF50;
|
||||||
|
border-radius: 3px;
|
||||||
|
}
|
||||||
|
|
||||||
|
QMessageBox QLabel {
|
||||||
|
color: #333333;
|
||||||
|
}
|
||||||
|
|
||||||
|
QMessageBox QPushButton {
|
||||||
|
min-width: 60px;
|
||||||
|
}
|
||||||
|
"""
|
||||||
|
|
||||||
|
# Color palette
|
||||||
|
COLORS = {
|
||||||
|
"primary": "#4CAF50",
|
||||||
|
"primary_hover": "#45a049",
|
||||||
|
"primary_active": "#3d8b40",
|
||||||
|
"background": "#f5f5f5",
|
||||||
|
"text": "#333333",
|
||||||
|
"text_secondary": "#666666",
|
||||||
|
"border": "#d0d0d0",
|
||||||
|
"success": "#4CAF50",
|
||||||
|
"error": "#f44336",
|
||||||
|
"warning": "#ff9800",
|
||||||
|
"info": "#2196F3",
|
||||||
|
}
|
||||||
|
|
||||||
|
# Font settings
|
||||||
|
FONTS = {
|
||||||
|
"default_size": 11,
|
||||||
|
"title_size": 16,
|
||||||
|
"mono_family": "Courier New",
|
||||||
|
}
|
||||||
|
|
||||||
|
|
||||||
|
def get_stylesheet() -> str:
|
||||||
|
"""Get the main stylesheet for the application"""
|
||||||
|
return MAIN_STYLESHEET
|
||||||
|
|
||||||
|
|
||||||
|
def get_color(color_name: str) -> str:
|
||||||
|
"""Get a color from the palette"""
|
||||||
|
return COLORS.get(color_name, "#000000")
|
||||||
1
farsi_transcriber/utils/__init__.py
Normal file
1
farsi_transcriber/utils/__init__.py
Normal file
@ -0,0 +1 @@
|
|||||||
|
"""Utility functions for Farsi Transcriber"""
|
||||||
164
farsi_transcriber/utils/export.py
Normal file
164
farsi_transcriber/utils/export.py
Normal file
@ -0,0 +1,164 @@
|
|||||||
|
"""
|
||||||
|
Export utilities for transcription results
|
||||||
|
|
||||||
|
Supports multiple export formats: TXT, SRT, JSON, TSV, VTT
|
||||||
|
"""
|
||||||
|
|
||||||
|
import json
|
||||||
|
from datetime import timedelta
|
||||||
|
from pathlib import Path
|
||||||
|
from typing import Dict, List
|
||||||
|
|
||||||
|
|
||||||
|
class TranscriptionExporter:
|
||||||
|
"""Export transcription results in various formats"""
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def export_txt(result: Dict, file_path: Path) -> None:
|
||||||
|
"""
|
||||||
|
Export transcription as plain text file.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Transcription result dictionary
|
||||||
|
file_path: Output file path
|
||||||
|
"""
|
||||||
|
text = result.get("full_text", "") or result.get("text", "")
|
||||||
|
|
||||||
|
with open(file_path, "w", encoding="utf-8") as f:
|
||||||
|
f.write(text)
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def export_srt(result: Dict, file_path: Path) -> None:
|
||||||
|
"""
|
||||||
|
Export transcription as SRT subtitle file.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Transcription result dictionary
|
||||||
|
file_path: Output file path
|
||||||
|
"""
|
||||||
|
segments = result.get("segments", [])
|
||||||
|
|
||||||
|
with open(file_path, "w", encoding="utf-8") as f:
|
||||||
|
for i, segment in enumerate(segments, 1):
|
||||||
|
start = TranscriptionExporter._format_srt_time(segment.get("start", 0))
|
||||||
|
end = TranscriptionExporter._format_srt_time(segment.get("end", 0))
|
||||||
|
text = segment.get("text", "").strip()
|
||||||
|
|
||||||
|
if text:
|
||||||
|
f.write(f"{i}\n")
|
||||||
|
f.write(f"{start} --> {end}\n")
|
||||||
|
f.write(f"{text}\n\n")
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def export_vtt(result: Dict, file_path: Path) -> None:
|
||||||
|
"""
|
||||||
|
Export transcription as WebVTT subtitle file.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Transcription result dictionary
|
||||||
|
file_path: Output file path
|
||||||
|
"""
|
||||||
|
segments = result.get("segments", [])
|
||||||
|
|
||||||
|
with open(file_path, "w", encoding="utf-8") as f:
|
||||||
|
f.write("WEBVTT\n\n")
|
||||||
|
|
||||||
|
for segment in segments:
|
||||||
|
start = TranscriptionExporter._format_vtt_time(segment.get("start", 0))
|
||||||
|
end = TranscriptionExporter._format_vtt_time(segment.get("end", 0))
|
||||||
|
text = segment.get("text", "").strip()
|
||||||
|
|
||||||
|
if text:
|
||||||
|
f.write(f"{start} --> {end}\n")
|
||||||
|
f.write(f"{text}\n\n")
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def export_json(result: Dict, file_path: Path) -> None:
|
||||||
|
"""
|
||||||
|
Export transcription as JSON file.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Transcription result dictionary
|
||||||
|
file_path: Output file path
|
||||||
|
"""
|
||||||
|
with open(file_path, "w", encoding="utf-8") as f:
|
||||||
|
json.dump(result, f, ensure_ascii=False, indent=2)
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def export_tsv(result: Dict, file_path: Path) -> None:
|
||||||
|
"""
|
||||||
|
Export transcription as TSV (tab-separated values) file.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Transcription result dictionary
|
||||||
|
file_path: Output file path
|
||||||
|
"""
|
||||||
|
segments = result.get("segments", [])
|
||||||
|
|
||||||
|
with open(file_path, "w", encoding="utf-8") as f:
|
||||||
|
# Write header
|
||||||
|
f.write("Index\tStart\tEnd\tDuration\tText\n")
|
||||||
|
|
||||||
|
for i, segment in enumerate(segments, 1):
|
||||||
|
start = segment.get("start", 0)
|
||||||
|
end = segment.get("end", 0)
|
||||||
|
duration = end - start
|
||||||
|
text = segment.get("text", "").strip()
|
||||||
|
|
||||||
|
if text:
|
||||||
|
f.write(
|
||||||
|
f"{i}\t{start:.2f}\t{end:.2f}\t{duration:.2f}\t{text}\n"
|
||||||
|
)
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def export(
|
||||||
|
result: Dict, file_path: Path, format_type: str = "txt"
|
||||||
|
) -> None:
|
||||||
|
"""
|
||||||
|
Export transcription in specified format.
|
||||||
|
|
||||||
|
Args:
|
||||||
|
result: Transcription result dictionary
|
||||||
|
file_path: Output file path
|
||||||
|
format_type: Export format ('txt', 'srt', 'vtt', 'json', 'tsv')
|
||||||
|
|
||||||
|
Raises:
|
||||||
|
ValueError: If format is not supported
|
||||||
|
"""
|
||||||
|
format_type = format_type.lower()
|
||||||
|
|
||||||
|
exporters = {
|
||||||
|
"txt": TranscriptionExporter.export_txt,
|
||||||
|
"srt": TranscriptionExporter.export_srt,
|
||||||
|
"vtt": TranscriptionExporter.export_vtt,
|
||||||
|
"json": TranscriptionExporter.export_json,
|
||||||
|
"tsv": TranscriptionExporter.export_tsv,
|
||||||
|
}
|
||||||
|
|
||||||
|
if format_type not in exporters:
|
||||||
|
raise ValueError(
|
||||||
|
f"Unsupported format: {format_type}. "
|
||||||
|
f"Supported formats: {list(exporters.keys())}"
|
||||||
|
)
|
||||||
|
|
||||||
|
exporters[format_type](result, file_path)
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def _format_srt_time(seconds: float) -> str:
|
||||||
|
"""Format time for SRT format (HH:MM:SS,mmm)"""
|
||||||
|
td = timedelta(seconds=seconds)
|
||||||
|
hours, remainder = divmod(int(td.total_seconds()), 3600)
|
||||||
|
minutes, secs = divmod(remainder, 60)
|
||||||
|
milliseconds = int((seconds % 1) * 1000)
|
||||||
|
|
||||||
|
return f"{hours:02d}:{minutes:02d}:{secs:02d},{milliseconds:03d}"
|
||||||
|
|
||||||
|
@staticmethod
|
||||||
|
def _format_vtt_time(seconds: float) -> str:
|
||||||
|
"""Format time for VTT format (HH:MM:SS.mmm)"""
|
||||||
|
td = timedelta(seconds=seconds)
|
||||||
|
hours, remainder = divmod(int(td.total_seconds()), 3600)
|
||||||
|
minutes, secs = divmod(remainder, 60)
|
||||||
|
milliseconds = int((seconds % 1) * 1000)
|
||||||
|
|
||||||
|
return f"{hours:02d}:{minutes:02d}:{secs:02d}.{milliseconds:03d}"
|
||||||
13
farsi_transcriber_web/.env.example
Normal file
13
farsi_transcriber_web/.env.example
Normal file
@ -0,0 +1,13 @@
|
|||||||
|
# Frontend environment variables
|
||||||
|
# Copy this to .env.local and update with your values
|
||||||
|
|
||||||
|
# API URL for the backend
|
||||||
|
# Local development: http://localhost:5000
|
||||||
|
# Railway production: https://your-backend-app.railway.app
|
||||||
|
VITE_API_URL=http://localhost:5000
|
||||||
|
|
||||||
|
# Application name
|
||||||
|
VITE_APP_NAME=Farsi Transcriber
|
||||||
|
|
||||||
|
# Max file size (in MB)
|
||||||
|
VITE_MAX_FILE_SIZE=500
|
||||||
7
farsi_transcriber_web/.env.production
Normal file
7
farsi_transcriber_web/.env.production
Normal file
@ -0,0 +1,7 @@
|
|||||||
|
# Production environment variables for Railway deployment
|
||||||
|
# Set VITE_API_URL in Railway environment variables instead of committing here
|
||||||
|
|
||||||
|
# Default fallback - will be overridden by Railway env var
|
||||||
|
VITE_API_URL=https://your-backend-url.railway.app
|
||||||
|
VITE_APP_NAME=Farsi Transcriber
|
||||||
|
VITE_MAX_FILE_SIZE=500
|
||||||
37
farsi_transcriber_web/.gitignore
vendored
Normal file
37
farsi_transcriber_web/.gitignore
vendored
Normal file
@ -0,0 +1,37 @@
|
|||||||
|
# Logs
|
||||||
|
logs
|
||||||
|
*.log
|
||||||
|
npm-debug.log*
|
||||||
|
yarn-debug.log*
|
||||||
|
yarn-error.log*
|
||||||
|
pnpm-debug.log*
|
||||||
|
lerna-debug.log*
|
||||||
|
|
||||||
|
node_modules
|
||||||
|
dist
|
||||||
|
dist-ssr
|
||||||
|
*.local
|
||||||
|
|
||||||
|
# Environment variables
|
||||||
|
.env
|
||||||
|
.env.local
|
||||||
|
.env.*.local
|
||||||
|
|
||||||
|
# Editor directories and files
|
||||||
|
.vscode
|
||||||
|
.idea
|
||||||
|
*.suo
|
||||||
|
*.ntvs*
|
||||||
|
*.njsproj
|
||||||
|
*.sln
|
||||||
|
*.sw?
|
||||||
|
|
||||||
|
# OS
|
||||||
|
.DS_Store
|
||||||
|
Thumbs.db
|
||||||
|
|
||||||
|
# Build output
|
||||||
|
*.tgz
|
||||||
|
*.tsbuildinfo
|
||||||
|
vite.config.js
|
||||||
|
vite.config.d.ts
|
||||||
308
farsi_transcriber_web/RAILWAY_DEPLOYMENT.md
Normal file
308
farsi_transcriber_web/RAILWAY_DEPLOYMENT.md
Normal file
@ -0,0 +1,308 @@
|
|||||||
|
# Railway Deployment Guide
|
||||||
|
|
||||||
|
Complete step-by-step guide to deploy your Farsi Transcriber web app to Railway.
|
||||||
|
|
||||||
|
## Prerequisites
|
||||||
|
|
||||||
|
1. **GitHub Account** - To connect your repository
|
||||||
|
2. **Railway Account** - Free signup at https://railway.app
|
||||||
|
3. **Git** - Already have this since you're using it
|
||||||
|
|
||||||
|
## Step 1: Prepare Your Code for Deployment
|
||||||
|
|
||||||
|
### 1.1 Make sure all files are committed
|
||||||
|
|
||||||
|
```bash
|
||||||
|
cd /home/user/whisper
|
||||||
|
git status
|
||||||
|
git add .
|
||||||
|
git commit -m "Ready for Railway deployment"
|
||||||
|
```
|
||||||
|
|
||||||
|
### 1.2 Push to GitHub
|
||||||
|
|
||||||
|
If you haven't already, push your fork to GitHub:
|
||||||
|
|
||||||
|
```bash
|
||||||
|
git push origin claude/review-repo-011CV3PVcA7ZSCTW2YquuMB8
|
||||||
|
```
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Step 2: Create Railway Account
|
||||||
|
|
||||||
|
1. Go to **https://railway.app**
|
||||||
|
2. Click **"Login with GitHub"**
|
||||||
|
3. Authorize Railway to access your GitHub account
|
||||||
|
4. You'll get **$5 monthly credit** for free ✅
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Step 3: Create Backend Service (Flask API)
|
||||||
|
|
||||||
|
### 3.1 Create a new project
|
||||||
|
|
||||||
|
1. In Railway dashboard, click **"Create New Project"**
|
||||||
|
2. Select **"GitHub Repo"**
|
||||||
|
3. Select your **whisper** repository
|
||||||
|
4. Railway will auto-detect it as Python
|
||||||
|
5. Configure:
|
||||||
|
- **Root Directory:** `farsi_transcriber_web/backend`
|
||||||
|
- **Start Command:** `gunicorn --workers 2 --bind 0.0.0.0:$PORT app:app`
|
||||||
|
|
||||||
|
### 3.2 Set environment variables
|
||||||
|
|
||||||
|
In the Railway dashboard for your backend:
|
||||||
|
|
||||||
|
1. Go to **Variables**
|
||||||
|
2. Add:
|
||||||
|
```
|
||||||
|
FLASK_ENV=production
|
||||||
|
FLASK_DEBUG=False
|
||||||
|
PYTHONUNBUFFERED=1
|
||||||
|
PORT=5000
|
||||||
|
```
|
||||||
|
|
||||||
|
### 3.3 Deploy
|
||||||
|
|
||||||
|
1. Click **Deploy**
|
||||||
|
2. Wait for deployment (takes 2-3 minutes)
|
||||||
|
3. Your backend URL will appear (e.g., `https://farsi-api-prod.railway.app`)
|
||||||
|
4. **Copy this URL** - you'll need it for the frontend
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Step 4: Create Frontend Service (React)
|
||||||
|
|
||||||
|
### 4.1 Create another service in same project
|
||||||
|
|
||||||
|
1. Go to your Railway project
|
||||||
|
2. Click **"Create New Service"**
|
||||||
|
3. Select **"GitHub Repo"**
|
||||||
|
4. Select your **whisper** repository
|
||||||
|
5. Configure:
|
||||||
|
- **Root Directory:** `farsi_transcriber_web`
|
||||||
|
- **Build Command:** `npm install && npm run build`
|
||||||
|
- **Start Command:** `npm run preview`
|
||||||
|
|
||||||
|
### 4.2 Set environment variables
|
||||||
|
|
||||||
|
In the Railway dashboard for your frontend:
|
||||||
|
|
||||||
|
1. Go to **Variables**
|
||||||
|
2. Add:
|
||||||
|
```
|
||||||
|
VITE_API_URL=https://your-backend-url-here.railway.app
|
||||||
|
VITE_APP_NAME=Farsi Transcriber
|
||||||
|
```
|
||||||
|
(Replace with your actual backend URL from Step 3.3)
|
||||||
|
|
||||||
|
### 4.3 Deploy
|
||||||
|
|
||||||
|
1. Click **Deploy**
|
||||||
|
2. Wait for deployment (2-5 minutes depending on npm install)
|
||||||
|
3. Your frontend URL will appear (e.g., `https://farsi-web-prod.railway.app`)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Step 5: Configure Services to Communicate
|
||||||
|
|
||||||
|
### 5.1 Link backend to frontend
|
||||||
|
|
||||||
|
1. In Railway dashboard, select frontend service
|
||||||
|
2. Go to **Variables**
|
||||||
|
3. Update `VITE_API_URL` with your backend service domain
|
||||||
|
4. Deploy again
|
||||||
|
|
||||||
|
### 5.2 Test the connection
|
||||||
|
|
||||||
|
1. Open your frontend URL
|
||||||
|
2. Try to add a file and transcribe
|
||||||
|
3. Check browser console for any errors
|
||||||
|
4. If errors, check Railway logs (click service → Logs)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Step 6: Monitor Your Deployment
|
||||||
|
|
||||||
|
### 6.1 View logs
|
||||||
|
|
||||||
|
In Railway dashboard:
|
||||||
|
- Click your service
|
||||||
|
- Go to **Logs** tab
|
||||||
|
- See real-time logs as users interact with your app
|
||||||
|
|
||||||
|
### 6.2 Check health
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# Check if backend is running
|
||||||
|
curl https://your-backend-url.railway.app/health
|
||||||
|
|
||||||
|
# Should return:
|
||||||
|
# {"status": "healthy", "model_loaded": true, "environment": "production"}
|
||||||
|
```
|
||||||
|
|
||||||
|
### 6.3 Monitor usage
|
||||||
|
|
||||||
|
- Railway dashboard shows RAM, CPU, bandwidth usage
|
||||||
|
- Your $5 credit should last 1-3 months for personal use
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Step 7: Custom Domain (Optional)
|
||||||
|
|
||||||
|
If you want a custom domain like `farsi.yourdomain.com`:
|
||||||
|
|
||||||
|
1. Buy a domain on GoDaddy, Namecheap, etc.
|
||||||
|
2. In Railway dashboard → Your app → Settings → Domains
|
||||||
|
3. Add custom domain
|
||||||
|
4. Update DNS records at your domain provider
|
||||||
|
5. Railway will handle SSL certificate automatically
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Troubleshooting
|
||||||
|
|
||||||
|
### Issue: Backend showing error "Model not loaded"
|
||||||
|
|
||||||
|
**Solution:** First transcription loads the 769MB model (takes 1-2 min). Wait and try again.
|
||||||
|
|
||||||
|
### Issue: Frontend can't reach backend
|
||||||
|
|
||||||
|
**Solution:**
|
||||||
|
1. Check backend URL is correct in frontend variables
|
||||||
|
2. Backend must be running (check Railway logs)
|
||||||
|
3. CORS should be enabled (already configured)
|
||||||
|
|
||||||
|
### Issue: Build fails
|
||||||
|
|
||||||
|
**Solution:**
|
||||||
|
1. Check Railway build logs for errors
|
||||||
|
2. Ensure `package.json` has all required dependencies
|
||||||
|
3. Run locally first: `npm install && npm run build`
|
||||||
|
|
||||||
|
### Issue: App runs slow
|
||||||
|
|
||||||
|
**Solution:**
|
||||||
|
1. You're on free tier with limited resources
|
||||||
|
2. Upgrade to paid tier ($5/month) for better performance
|
||||||
|
3. Or wait for model to cache (subsequent transcriptions are fast)
|
||||||
|
|
||||||
|
### Issue: Out of memory
|
||||||
|
|
||||||
|
**Solution:**
|
||||||
|
1. Free tier has limited RAM
|
||||||
|
2. Close unused tabs/apps
|
||||||
|
3. Use smaller Whisper model (edit backend to use 'small' instead of 'medium')
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Next Steps: Custom Domain Setup
|
||||||
|
|
||||||
|
Once stable, add your custom domain:
|
||||||
|
|
||||||
|
1. Purchase domain
|
||||||
|
2. Railway → Settings → Domains → Add Domain
|
||||||
|
3. Update DNS CNAME records
|
||||||
|
4. Railway auto-generates SSL certificate
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Cost Breakdown
|
||||||
|
|
||||||
|
### Free Tier ($5/month credit)
|
||||||
|
|
||||||
|
- ✅ 500 build minutes/month
|
||||||
|
- ✅ 100 GB bandwidth/month
|
||||||
|
- ✅ 6,000 compute unit hours
|
||||||
|
- ✅ More than enough for personal use
|
||||||
|
|
||||||
|
### Your app will cost:
|
||||||
|
|
||||||
|
- **Backend (Flask):** ~$1-2/month
|
||||||
|
- **Frontend (React):** ~$0.50/month
|
||||||
|
- **Total:** ~$2/month (with free credit covering 2-3 months)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Useful Commands
|
||||||
|
|
||||||
|
### Check if Railway CLI is installed
|
||||||
|
|
||||||
|
```bash
|
||||||
|
railway --version
|
||||||
|
```
|
||||||
|
|
||||||
|
### Install Railway CLI
|
||||||
|
|
||||||
|
```bash
|
||||||
|
npm i -g @railway/cli
|
||||||
|
```
|
||||||
|
|
||||||
|
### Deploy from command line
|
||||||
|
|
||||||
|
```bash
|
||||||
|
railway up
|
||||||
|
```
|
||||||
|
|
||||||
|
### View logs
|
||||||
|
|
||||||
|
```bash
|
||||||
|
railway logs
|
||||||
|
```
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## What Happens Now
|
||||||
|
|
||||||
|
1. ✅ Your app is live on Railway
|
||||||
|
2. ✅ Free $5 monthly credit
|
||||||
|
3. ✅ Auto-scaling (if you get traffic)
|
||||||
|
4. ✅ 24/7 uptime
|
||||||
|
5. ✅ Automatic SSL/HTTPS
|
||||||
|
6. ✅ No infrastructure to manage
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Monitor Your App
|
||||||
|
|
||||||
|
Visit your Railway dashboard regularly to:
|
||||||
|
- Check resource usage
|
||||||
|
- View logs
|
||||||
|
- Update environment variables
|
||||||
|
- Scale services if needed
|
||||||
|
- Monitor costs
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## After Deployment
|
||||||
|
|
||||||
|
Your app is now online! Share the URL with friends:
|
||||||
|
|
||||||
|
```
|
||||||
|
https://your-app-name.railway.app
|
||||||
|
```
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Further Reading
|
||||||
|
|
||||||
|
- [Railway Documentation](https://docs.railway.app)
|
||||||
|
- [Railway GitHub Integration](https://docs.railway.app/guides/github)
|
||||||
|
- [Railway Environment Variables](https://docs.railway.app/develop/variables)
|
||||||
|
- [Whisper API Docs](https://github.com/openai/whisper)
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
## Support
|
||||||
|
|
||||||
|
If you have issues:
|
||||||
|
|
||||||
|
1. Check Railway logs (click service → Logs)
|
||||||
|
2. Check browser console (F12 → Console tab)
|
||||||
|
3. Visit Railway docs: https://docs.railway.app
|
||||||
|
4. Check Flask logs for backend errors
|
||||||
|
|
||||||
|
---
|
||||||
|
|
||||||
|
**Congratulations! Your Farsi Transcriber is now live!** 🎉
|
||||||
384
farsi_transcriber_web/README.md
Normal file
384
farsi_transcriber_web/README.md
Normal file
@ -0,0 +1,384 @@
|
|||||||
|
# Farsi Transcriber - Web Application
|
||||||
|
|
||||||
|
A professional web-based application for transcribing Farsi audio and video files using OpenAI's Whisper model.
|
||||||
|
|
||||||
|
## Features
|
||||||
|
|
||||||
|
✨ **Core Features**
|
||||||
|
- 🎙️ Transcribe audio files (MP3, WAV, M4A, FLAC, OGG, AAC, WMA)
|
||||||
|
- 🎬 Extract audio from video files (MP4, MKV, MOV, WebM, AVI, FLV, WMV)
|
||||||
|
- 🇮🇷 High-accuracy Farsi/Persian language transcription
|
||||||
|
- ⏱️ Word-level timestamps for precise timing
|
||||||
|
- 📤 Export to multiple formats (TXT, SRT, VTT, JSON)
|
||||||
|
- 💻 Clean, intuitive React-based UI with Figma design
|
||||||
|
- 🎨 Dark/Light theme toggle
|
||||||
|
- 🔍 Search and text highlighting in transcriptions
|
||||||
|
- 📋 File queue management
|
||||||
|
- 💾 Copy individual transcription segments
|
||||||
|
- 🚀 GPU acceleration support (CUDA)
|
||||||
|
- 🎯 Resizable window for flexible workspace
|
||||||
|
|
||||||
|
## Tech Stack
|
||||||
|
|
||||||
|
**Frontend:**
|
||||||
|
- React 18+ with TypeScript
|
||||||
|
- Vite (fast build tool)
|
||||||
|
- Tailwind CSS v4.0
|
||||||
|
- Lucide React (icons)
|
||||||
|
- re-resizable (window resizing)
|
||||||
|
- Sonner (toast notifications)
|
||||||
|
|
||||||
|
**Backend:**
|
||||||
|
- Flask (Python web framework)
|
||||||
|
- OpenAI Whisper (speech recognition)
|
||||||
|
- PyTorch (deep learning)
|
||||||
|
- Flask-CORS (cross-origin requests)
|
||||||
|
|
||||||
|
## System Requirements
|
||||||
|
|
||||||
|
**Frontend:**
|
||||||
|
- Node.js 16+
|
||||||
|
- npm/yarn/pnpm
|
||||||
|
|
||||||
|
**Backend:**
|
||||||
|
- Python 3.8+
|
||||||
|
- 4GB RAM minimum
|
||||||
|
- 8GB+ recommended
|
||||||
|
- ffmpeg installed
|
||||||
|
- Optional: NVIDIA GPU with CUDA support
|
||||||
|
|
||||||
|
## Installation
|
||||||
|
|
||||||
|
### Step 1: Install ffmpeg
|
||||||
|
|
||||||
|
Choose your operating system:
|
||||||
|
|
||||||
|
**Ubuntu/Debian:**
|
||||||
|
```bash
|
||||||
|
sudo apt update && sudo apt install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
**macOS (Homebrew):**
|
||||||
|
```bash
|
||||||
|
brew install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
**Windows (Chocolatey):**
|
||||||
|
```bash
|
||||||
|
choco install ffmpeg
|
||||||
|
```
|
||||||
|
|
||||||
|
### Step 2: Backend Setup
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# Navigate to backend directory
|
||||||
|
cd backend
|
||||||
|
|
||||||
|
# Create virtual environment
|
||||||
|
python3 -m venv venv
|
||||||
|
source venv/bin/activate # On Windows: venv\Scripts\activate
|
||||||
|
|
||||||
|
# Install dependencies
|
||||||
|
pip install -r requirements.txt
|
||||||
|
```
|
||||||
|
|
||||||
|
### Step 3: Frontend Setup
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# Navigate to root directory
|
||||||
|
cd ..
|
||||||
|
|
||||||
|
# Install Node dependencies
|
||||||
|
npm install
|
||||||
|
|
||||||
|
# Or use yarn/pnpm
|
||||||
|
yarn install
|
||||||
|
# or
|
||||||
|
pnpm install
|
||||||
|
```
|
||||||
|
|
||||||
|
## Running the Application
|
||||||
|
|
||||||
|
### Step 1: Start Backend API
|
||||||
|
|
||||||
|
```bash
|
||||||
|
cd backend
|
||||||
|
source venv/bin/activate # Activate virtual environment
|
||||||
|
python app.py
|
||||||
|
```
|
||||||
|
|
||||||
|
The API will be available at `http://localhost:5000`
|
||||||
|
|
||||||
|
### Step 2: Start Frontend Dev Server
|
||||||
|
|
||||||
|
In a new terminal:
|
||||||
|
|
||||||
|
```bash
|
||||||
|
npm run dev
|
||||||
|
```
|
||||||
|
|
||||||
|
The application will be available at `http://localhost:3000`
|
||||||
|
|
||||||
|
## Building for Production
|
||||||
|
|
||||||
|
### Frontend Build
|
||||||
|
|
||||||
|
```bash
|
||||||
|
npm run build
|
||||||
|
```
|
||||||
|
|
||||||
|
This creates optimized production build in `dist/` directory.
|
||||||
|
|
||||||
|
### Backend Deployment
|
||||||
|
|
||||||
|
For production, use a production WSGI server:
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# Install Gunicorn
|
||||||
|
pip install gunicorn
|
||||||
|
|
||||||
|
# Run with Gunicorn
|
||||||
|
gunicorn -w 4 -b 0.0.0.0:5000 app:app
|
||||||
|
```
|
||||||
|
|
||||||
|
## API Endpoints
|
||||||
|
|
||||||
|
### `/health` (GET)
|
||||||
|
Health check endpoint
|
||||||
|
|
||||||
|
**Response:**
|
||||||
|
```json
|
||||||
|
{
|
||||||
|
"status": "healthy",
|
||||||
|
"model_loaded": true,
|
||||||
|
"device": "cuda|cpu"
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
### `/transcribe` (POST)
|
||||||
|
Transcribe audio/video file
|
||||||
|
|
||||||
|
**Request:**
|
||||||
|
- `file`: Audio/video file (multipart/form-data)
|
||||||
|
- `language`: Language code (optional, default: "fa" for Farsi)
|
||||||
|
|
||||||
|
**Response:**
|
||||||
|
```json
|
||||||
|
{
|
||||||
|
"status": "success",
|
||||||
|
"filename": "audio.mp3",
|
||||||
|
"language": "fa",
|
||||||
|
"text": "Full transcription text...",
|
||||||
|
"segments": [
|
||||||
|
{
|
||||||
|
"start": "00:00:00.000",
|
||||||
|
"end": "00:00:05.500",
|
||||||
|
"text": "سلام دنیا"
|
||||||
|
}
|
||||||
|
]
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
### `/models` (GET)
|
||||||
|
Get available Whisper models
|
||||||
|
|
||||||
|
**Response:**
|
||||||
|
```json
|
||||||
|
{
|
||||||
|
"available_models": ["tiny", "base", "small", "medium", "large"],
|
||||||
|
"current_model": "medium",
|
||||||
|
"description": "..."
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
### `/export` (POST)
|
||||||
|
Export transcription
|
||||||
|
|
||||||
|
**Request:**
|
||||||
|
```json
|
||||||
|
{
|
||||||
|
"transcription": "Full text...",
|
||||||
|
"segments": [...],
|
||||||
|
"format": "txt|srt|vtt|json"
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
**Response:**
|
||||||
|
```json
|
||||||
|
{
|
||||||
|
"status": "success",
|
||||||
|
"format": "srt",
|
||||||
|
"content": "...",
|
||||||
|
"mime_type": "text/plain"
|
||||||
|
}
|
||||||
|
```
|
||||||
|
|
||||||
|
## Usage Guide
|
||||||
|
|
||||||
|
### 1. Add Files to Queue
|
||||||
|
- Click "Add Files" button in the left sidebar
|
||||||
|
- Select audio or video files
|
||||||
|
- Multiple files can be added to the queue
|
||||||
|
|
||||||
|
### 2. Transcribe
|
||||||
|
- Select a file from the queue
|
||||||
|
- Click "Transcribe" button
|
||||||
|
- Watch the progress indicator
|
||||||
|
- Results appear with timestamps
|
||||||
|
|
||||||
|
### 3. Search & Copy
|
||||||
|
- Use the search bar to find specific text
|
||||||
|
- Matching text is highlighted
|
||||||
|
- Click copy icon to copy individual segments
|
||||||
|
|
||||||
|
### 4. Export Results
|
||||||
|
- Select export format (TXT, SRT, VTT, JSON)
|
||||||
|
- Click "Export" button
|
||||||
|
- File is downloaded or ready to save
|
||||||
|
|
||||||
|
### 5. Theme Toggle
|
||||||
|
- Click sun/moon icon in header
|
||||||
|
- Switch between light and dark themes
|
||||||
|
|
||||||
|
## Project Structure
|
||||||
|
|
||||||
|
```
|
||||||
|
farsi_transcriber_web/
|
||||||
|
├── src/
|
||||||
|
│ ├── App.tsx # Main application component
|
||||||
|
│ ├── main.tsx # React entry point
|
||||||
|
│ ├── index.css # Global styles
|
||||||
|
│ └── components/
|
||||||
|
│ ├── Button.tsx
|
||||||
|
│ ├── Progress.tsx
|
||||||
|
│ ├── Input.tsx
|
||||||
|
│ └── Select.tsx
|
||||||
|
├── backend/
|
||||||
|
│ ├── app.py # Flask API server
|
||||||
|
│ ├── requirements.txt # Python dependencies
|
||||||
|
│ └── .gitignore
|
||||||
|
├── public/
|
||||||
|
├── package.json
|
||||||
|
├── vite.config.ts
|
||||||
|
├── tsconfig.json
|
||||||
|
├── tailwind.config.js
|
||||||
|
├── postcss.config.js
|
||||||
|
└── README.md
|
||||||
|
```
|
||||||
|
|
||||||
|
## Configuration
|
||||||
|
|
||||||
|
### Environment Variables
|
||||||
|
|
||||||
|
Create a `.env.local` file in the root directory:
|
||||||
|
|
||||||
|
```
|
||||||
|
VITE_API_URL=http://localhost:5000
|
||||||
|
VITE_MAX_FILE_SIZE=500MB
|
||||||
|
```
|
||||||
|
|
||||||
|
### Backend Configuration
|
||||||
|
|
||||||
|
Edit `backend/app.py` to customize:
|
||||||
|
|
||||||
|
```python
|
||||||
|
# Change model size
|
||||||
|
model = whisper.load_model('large') # tiny, base, small, medium, large
|
||||||
|
|
||||||
|
# Change upload folder
|
||||||
|
UPLOAD_FOLDER = '/custom/path'
|
||||||
|
|
||||||
|
# Change max file size
|
||||||
|
MAX_FILE_SIZE = 1024 * 1024 * 1024 # 1GB
|
||||||
|
```
|
||||||
|
|
||||||
|
## Troubleshooting
|
||||||
|
|
||||||
|
### Issue: "API connection failed"
|
||||||
|
**Solution**: Ensure backend is running on `http://localhost:5000`
|
||||||
|
|
||||||
|
### Issue: "Whisper model not found"
|
||||||
|
**Solution**: First run downloads the model (~3GB). Ensure internet connection and disk space.
|
||||||
|
|
||||||
|
### Issue: "CUDA out of memory"
|
||||||
|
**Solution**: Use smaller model or reduce batch size in `backend/app.py`
|
||||||
|
|
||||||
|
### Issue: "ffmpeg not found"
|
||||||
|
**Solution**: Install ffmpeg using your package manager (see Installation section)
|
||||||
|
|
||||||
|
### Issue: Port 3000 or 5000 already in use
|
||||||
|
**Solution**: Change ports in `vite.config.ts` and `backend/app.py`
|
||||||
|
|
||||||
|
## Performance Tips
|
||||||
|
|
||||||
|
1. **Use GPU** - Ensure NVIDIA CUDA is properly installed
|
||||||
|
2. **Choose appropriate model** - Balance speed vs accuracy
|
||||||
|
3. **Close other applications** - Free up RAM/VRAM
|
||||||
|
4. **Use SSD** - Faster model loading and file I/O
|
||||||
|
5. **Batch Processing** - Process multiple files sequentially
|
||||||
|
|
||||||
|
## Future Enhancements
|
||||||
|
|
||||||
|
- [ ] Drag-and-drop file upload
|
||||||
|
- [ ] Audio playback synchronized with transcription
|
||||||
|
- [ ] Edit segments inline
|
||||||
|
- [ ] Keyboard shortcuts
|
||||||
|
- [ ] Save/load sessions
|
||||||
|
- [ ] Speaker diarization
|
||||||
|
- [ ] Confidence scores
|
||||||
|
- [ ] Custom vocabulary support
|
||||||
|
|
||||||
|
## Development
|
||||||
|
|
||||||
|
### Code Style
|
||||||
|
|
||||||
|
```bash
|
||||||
|
# Format code (if ESLint configured)
|
||||||
|
npm run lint
|
||||||
|
|
||||||
|
# Build for development
|
||||||
|
npm run dev
|
||||||
|
|
||||||
|
# Build for production
|
||||||
|
npm run build
|
||||||
|
```
|
||||||
|
|
||||||
|
### Adding Components
|
||||||
|
|
||||||
|
New components go in `src/components/` and should:
|
||||||
|
- Use TypeScript
|
||||||
|
- Include prop interfaces
|
||||||
|
- Export as default
|
||||||
|
- Include JSDoc comments
|
||||||
|
|
||||||
|
## Common Issues & Solutions
|
||||||
|
|
||||||
|
| Issue | Solution |
|
||||||
|
|-------|----------|
|
||||||
|
| Models slow to load | GPU required for fast transcription |
|
||||||
|
| File not supported | Check file extension is in supported list |
|
||||||
|
| Transcription has errors | Try larger model (medium/large) |
|
||||||
|
| Application crashes | Check browser console and Flask logs |
|
||||||
|
| Export not working | Ensure segments data is complete |
|
||||||
|
|
||||||
|
## License
|
||||||
|
|
||||||
|
MIT License - Personal use and modifications allowed
|
||||||
|
|
||||||
|
## Credits
|
||||||
|
|
||||||
|
Built with:
|
||||||
|
- [OpenAI Whisper](https://github.com/openai/whisper) - Speech recognition
|
||||||
|
- [React](https://react.dev/) - UI framework
|
||||||
|
- [Vite](https://vitejs.dev/) - Build tool
|
||||||
|
- [Tailwind CSS](https://tailwindcss.com/) - Styling
|
||||||
|
- [Flask](https://flask.palletsprojects.com/) - Backend framework
|
||||||
|
|
||||||
|
## Support
|
||||||
|
|
||||||
|
For issues:
|
||||||
|
1. Check the troubleshooting section
|
||||||
|
2. Verify ffmpeg is installed
|
||||||
|
3. Check Flask backend logs
|
||||||
|
4. Review browser console for errors
|
||||||
|
5. Ensure Python 3.8+ and Node.js 16+ are installed
|
||||||
21
farsi_transcriber_web/backend/.env.example
Normal file
21
farsi_transcriber_web/backend/.env.example
Normal file
@ -0,0 +1,21 @@
|
|||||||
|
# Backend environment variables
|
||||||
|
# Copy this to .env and update with your values
|
||||||
|
|
||||||
|
# Flask environment
|
||||||
|
FLASK_ENV=production
|
||||||
|
FLASK_DEBUG=False
|
||||||
|
|
||||||
|
# Server port (Railway sets PORT automatically)
|
||||||
|
PORT=5000
|
||||||
|
|
||||||
|
# CORS settings
|
||||||
|
FLASK_CORS_ORIGINS=*
|
||||||
|
|
||||||
|
# Whisper model settings
|
||||||
|
WHISPER_MODEL=medium
|
||||||
|
|
||||||
|
# File upload settings
|
||||||
|
MAX_FILE_SIZE=500000000
|
||||||
|
|
||||||
|
# Python path
|
||||||
|
PYTHONUNBUFFERED=1
|
||||||
42
farsi_transcriber_web/backend/.gitignore
vendored
Normal file
42
farsi_transcriber_web/backend/.gitignore
vendored
Normal file
@ -0,0 +1,42 @@
|
|||||||
|
# Python
|
||||||
|
__pycache__/
|
||||||
|
*.py[cod]
|
||||||
|
*$py.class
|
||||||
|
*.so
|
||||||
|
.Python
|
||||||
|
build/
|
||||||
|
develop-eggs/
|
||||||
|
dist/
|
||||||
|
downloads/
|
||||||
|
eggs/
|
||||||
|
.eggs/
|
||||||
|
lib/
|
||||||
|
lib64/
|
||||||
|
parts/
|
||||||
|
sdist/
|
||||||
|
var/
|
||||||
|
wheels/
|
||||||
|
*.egg-info/
|
||||||
|
.installed.cfg
|
||||||
|
*.egg
|
||||||
|
|
||||||
|
# Virtual Environment
|
||||||
|
venv/
|
||||||
|
ENV/
|
||||||
|
env/
|
||||||
|
|
||||||
|
# IDE
|
||||||
|
.vscode/
|
||||||
|
.idea/
|
||||||
|
*.swp
|
||||||
|
*.swo
|
||||||
|
*~
|
||||||
|
|
||||||
|
# ML Models
|
||||||
|
*.pt
|
||||||
|
*.pth
|
||||||
|
~/.cache/whisper/
|
||||||
|
|
||||||
|
# Uploads
|
||||||
|
/uploads
|
||||||
|
/tmp
|
||||||
1
farsi_transcriber_web/backend/Procfile
Normal file
1
farsi_transcriber_web/backend/Procfile
Normal file
@ -0,0 +1 @@
|
|||||||
|
web: gunicorn --workers 4 --worker-class sync --bind 0.0.0.0:$PORT app:app
|
||||||
228
farsi_transcriber_web/backend/app.py
Normal file
228
farsi_transcriber_web/backend/app.py
Normal file
@ -0,0 +1,228 @@
|
|||||||
|
"""
|
||||||
|
Farsi Transcriber Backend API
|
||||||
|
|
||||||
|
Flask API for handling audio/video file transcription using Whisper model.
|
||||||
|
Configured for Railway deployment with lazy model loading.
|
||||||
|
"""
|
||||||
|
|
||||||
|
import os
|
||||||
|
import sys
|
||||||
|
import tempfile
|
||||||
|
from pathlib import Path
|
||||||
|
from werkzeug.utils import secure_filename
|
||||||
|
from flask import Flask, request, jsonify
|
||||||
|
from flask_cors import CORS
|
||||||
|
|
||||||
|
# Prevent model download during build
|
||||||
|
os.environ['WHISPER_CACHE'] = os.path.expanduser('~/.cache/whisper')
|
||||||
|
|
||||||
|
# Add parent directory to path for imports
|
||||||
|
sys.path.insert(0, str(Path(__file__).parent.parent.parent))
|
||||||
|
|
||||||
|
app = Flask(__name__)
|
||||||
|
CORS(app, resources={r"/api/*": {"origins": "*"}})
|
||||||
|
|
||||||
|
# Configuration
|
||||||
|
UPLOAD_FOLDER = tempfile.gettempdir()
|
||||||
|
ALLOWED_EXTENSIONS = {'mp3', 'wav', 'm4a', 'flac', 'ogg', 'aac', 'wma', 'mp4', 'mkv', 'mov', 'webm', 'avi', 'flv', 'wmv'}
|
||||||
|
MAX_FILE_SIZE = 500 * 1024 * 1024 # 500MB
|
||||||
|
|
||||||
|
# Production settings
|
||||||
|
app.config['UPLOAD_FOLDER'] = UPLOAD_FOLDER
|
||||||
|
app.config['MAX_CONTENT_LENGTH'] = MAX_FILE_SIZE
|
||||||
|
app.config['ENV'] = os.getenv('FLASK_ENV', 'production')
|
||||||
|
|
||||||
|
# Load Whisper model (lazy load - only on first transcription request)
|
||||||
|
model = None
|
||||||
|
|
||||||
|
def load_model():
|
||||||
|
"""Lazy load Whisper model on first use (not during build)"""
|
||||||
|
global model
|
||||||
|
if model is None:
|
||||||
|
try:
|
||||||
|
print("⏳ Loading Whisper model for first time...")
|
||||||
|
print(" This may take 1-2 minutes on first run...")
|
||||||
|
# Import here to avoid loading during build
|
||||||
|
import whisper
|
||||||
|
model = whisper.load_model('medium')
|
||||||
|
print("✓ Whisper model loaded successfully")
|
||||||
|
except Exception as e:
|
||||||
|
print(f"✗ Error loading Whisper model: {e}")
|
||||||
|
model = None
|
||||||
|
return model
|
||||||
|
|
||||||
|
|
||||||
|
def allowed_file(filename):
|
||||||
|
"""Check if file has allowed extension"""
|
||||||
|
return '.' in filename and filename.rsplit('.', 1)[1].lower() in ALLOWED_EXTENSIONS
|
||||||
|
|
||||||
|
|
||||||
|
@app.route('/', methods=['GET'])
|
||||||
|
def index():
|
||||||
|
"""Root endpoint"""
|
||||||
|
return jsonify({
|
||||||
|
'message': 'Farsi Transcriber API',
|
||||||
|
'version': '1.0.0',
|
||||||
|
'status': 'running'
|
||||||
|
})
|
||||||
|
|
||||||
|
|
||||||
|
@app.route('/health', methods=['GET'])
|
||||||
|
def health():
|
||||||
|
"""Health check endpoint - fast response without loading model"""
|
||||||
|
return jsonify({
|
||||||
|
'status': 'healthy',
|
||||||
|
'model_loaded': model is not None,
|
||||||
|
'environment': app.config['ENV']
|
||||||
|
})
|
||||||
|
|
||||||
|
|
||||||
|
@app.route('/transcribe', methods=['POST'])
|
||||||
|
def transcribe():
|
||||||
|
"""
|
||||||
|
Transcribe audio/video file
|
||||||
|
|
||||||
|
Request:
|
||||||
|
- file: Audio/video file
|
||||||
|
- language: Language code (default: 'fa' for Farsi)
|
||||||
|
|
||||||
|
Response:
|
||||||
|
- transcription results with segments and timestamps
|
||||||
|
"""
|
||||||
|
try:
|
||||||
|
# Load model if not already loaded
|
||||||
|
whisper_model = load_model()
|
||||||
|
if not whisper_model:
|
||||||
|
return jsonify({'error': 'Failed to load Whisper model'}), 500
|
||||||
|
|
||||||
|
# Check if file is in request
|
||||||
|
if 'file' not in request.files:
|
||||||
|
return jsonify({'error': 'No file provided'}), 400
|
||||||
|
|
||||||
|
file = request.files['file']
|
||||||
|
|
||||||
|
if file.filename == '':
|
||||||
|
return jsonify({'error': 'No file selected'}), 400
|
||||||
|
|
||||||
|
if not allowed_file(file.filename):
|
||||||
|
return jsonify({'error': 'File type not allowed'}), 400
|
||||||
|
|
||||||
|
# Save file
|
||||||
|
filename = secure_filename(file.filename)
|
||||||
|
filepath = os.path.join(app.config['UPLOAD_FOLDER'], filename)
|
||||||
|
file.save(filepath)
|
||||||
|
|
||||||
|
# Get language code from request (default: Farsi)
|
||||||
|
language = request.form.get('language', 'fa')
|
||||||
|
|
||||||
|
# Transcribe
|
||||||
|
result = whisper_model.transcribe(filepath, language=language, verbose=False)
|
||||||
|
|
||||||
|
# Format response
|
||||||
|
segments = []
|
||||||
|
for segment in result.get('segments', []):
|
||||||
|
segments.append({
|
||||||
|
'start': f"{int(segment['start'] // 3600):02d}:{int((segment['start'] % 3600) // 60):02d}:{int(segment['start'] % 60):02d}.{int((segment['start'] % 1) * 1000):03d}",
|
||||||
|
'end': f"{int(segment['end'] // 3600):02d}:{int((segment['end'] % 3600) // 60):02d}:{int(segment['end'] % 60):02d}.{int((segment['end'] % 1) * 1000):03d}",
|
||||||
|
'text': segment['text'].strip(),
|
||||||
|
})
|
||||||
|
|
||||||
|
# Clean up uploaded file
|
||||||
|
try:
|
||||||
|
os.remove(filepath)
|
||||||
|
except:
|
||||||
|
pass
|
||||||
|
|
||||||
|
return jsonify({
|
||||||
|
'status': 'success',
|
||||||
|
'filename': filename,
|
||||||
|
'language': result.get('language', 'unknown'),
|
||||||
|
'text': result.get('text', ''),
|
||||||
|
'segments': segments
|
||||||
|
})
|
||||||
|
|
||||||
|
except Exception as e:
|
||||||
|
return jsonify({'error': str(e)}), 500
|
||||||
|
|
||||||
|
|
||||||
|
@app.route('/models', methods=['GET'])
|
||||||
|
def get_models():
|
||||||
|
"""Get available Whisper models"""
|
||||||
|
return jsonify({
|
||||||
|
'available_models': ['tiny', 'base', 'small', 'medium', 'large'],
|
||||||
|
'current_model': 'medium',
|
||||||
|
'description': 'List of available Whisper models. Larger models are more accurate but slower.'
|
||||||
|
})
|
||||||
|
|
||||||
|
|
||||||
|
@app.route('/export', methods=['POST'])
|
||||||
|
def export():
|
||||||
|
"""
|
||||||
|
Export transcription in specified format
|
||||||
|
|
||||||
|
Request:
|
||||||
|
- transcription: Full transcription text
|
||||||
|
- segments: Array of segments with timestamps
|
||||||
|
- format: Export format (txt, srt, vtt, json)
|
||||||
|
|
||||||
|
Response:
|
||||||
|
- Exported file content
|
||||||
|
"""
|
||||||
|
try:
|
||||||
|
data = request.json
|
||||||
|
transcription = data.get('transcription', '')
|
||||||
|
segments = data.get('segments', [])
|
||||||
|
format_type = data.get('format', 'txt').lower()
|
||||||
|
|
||||||
|
if format_type == 'txt':
|
||||||
|
content = transcription
|
||||||
|
mime_type = 'text/plain'
|
||||||
|
elif format_type == 'srt':
|
||||||
|
content = _format_srt(segments)
|
||||||
|
mime_type = 'text/plain'
|
||||||
|
elif format_type == 'vtt':
|
||||||
|
content = _format_vtt(segments)
|
||||||
|
mime_type = 'text/plain'
|
||||||
|
elif format_type == 'json':
|
||||||
|
import json
|
||||||
|
content = json.dumps({'text': transcription, 'segments': segments}, ensure_ascii=False, indent=2)
|
||||||
|
mime_type = 'application/json'
|
||||||
|
else:
|
||||||
|
return jsonify({'error': 'Unsupported format'}), 400
|
||||||
|
|
||||||
|
return jsonify({
|
||||||
|
'status': 'success',
|
||||||
|
'format': format_type,
|
||||||
|
'content': content,
|
||||||
|
'mime_type': mime_type
|
||||||
|
})
|
||||||
|
|
||||||
|
except Exception as e:
|
||||||
|
return jsonify({'error': str(e)}), 500
|
||||||
|
|
||||||
|
|
||||||
|
def _format_srt(segments):
|
||||||
|
"""Format transcription as SRT subtitle format"""
|
||||||
|
lines = []
|
||||||
|
for i, segment in enumerate(segments, 1):
|
||||||
|
lines.append(str(i))
|
||||||
|
lines.append(f"{segment['start']} --> {segment['end']}")
|
||||||
|
lines.append(segment['text'])
|
||||||
|
lines.append('')
|
||||||
|
return '\n'.join(lines)
|
||||||
|
|
||||||
|
|
||||||
|
def _format_vtt(segments):
|
||||||
|
"""Format transcription as WebVTT subtitle format"""
|
||||||
|
lines = ['WEBVTT', '']
|
||||||
|
for segment in segments:
|
||||||
|
lines.append(f"{segment['start']} --> {segment['end']}")
|
||||||
|
lines.append(segment['text'])
|
||||||
|
lines.append('')
|
||||||
|
return '\n'.join(lines)
|
||||||
|
|
||||||
|
|
||||||
|
if __name__ == '__main__':
|
||||||
|
port = int(os.getenv('PORT', 5000))
|
||||||
|
debug = os.getenv('FLASK_ENV', 'production') == 'development'
|
||||||
|
app.run(debug=debug, host='0.0.0.0', port=port, threaded=True)
|
||||||
11
farsi_transcriber_web/backend/nixpacks.toml
Normal file
11
farsi_transcriber_web/backend/nixpacks.toml
Normal file
@ -0,0 +1,11 @@
|
|||||||
|
# Backend Nixpacks configuration
|
||||||
|
# Ensures ffmpeg is available for Whisper audio processing
|
||||||
|
|
||||||
|
[phases.setup]
|
||||||
|
nixPkgs = ["ffmpeg"]
|
||||||
|
|
||||||
|
[phases.install]
|
||||||
|
cmds = ["pip install -r requirements.txt"]
|
||||||
|
|
||||||
|
[start]
|
||||||
|
cmd = "gunicorn --workers 2 --worker-class sync --timeout 120 --bind 0.0.0.0:$PORT app:app"
|
||||||
10
farsi_transcriber_web/backend/requirements.txt
Normal file
10
farsi_transcriber_web/backend/requirements.txt
Normal file
@ -0,0 +1,10 @@
|
|||||||
|
Flask==2.3.3
|
||||||
|
Flask-CORS==4.0.0
|
||||||
|
python-dotenv==1.0.0
|
||||||
|
openai-whisper>=20230314
|
||||||
|
torch>=1.10.1
|
||||||
|
numpy>=1.21.0
|
||||||
|
python-multipart==0.0.6
|
||||||
|
gunicorn==21.2.0
|
||||||
|
|
||||||
|
|
||||||
13
farsi_transcriber_web/index.html
Normal file
13
farsi_transcriber_web/index.html
Normal file
@ -0,0 +1,13 @@
|
|||||||
|
<!doctype html>
|
||||||
|
<html lang="en">
|
||||||
|
<head>
|
||||||
|
<meta charset="UTF-8" />
|
||||||
|
<link rel="icon" type="image/svg+xml" href="/vite.svg" />
|
||||||
|
<meta name="viewport" content="width=device-width, initial-scale=1.0" />
|
||||||
|
<title>Farsi Audio/Video Transcriber</title>
|
||||||
|
</head>
|
||||||
|
<body>
|
||||||
|
<div id="root"></div>
|
||||||
|
<script type="module" src="/src/main.tsx"></script>
|
||||||
|
</body>
|
||||||
|
</html>
|
||||||
11
farsi_transcriber_web/nixpacks.toml
Normal file
11
farsi_transcriber_web/nixpacks.toml
Normal file
@ -0,0 +1,11 @@
|
|||||||
|
# Frontend Nixpacks configuration
|
||||||
|
# Node.js React app with Vite
|
||||||
|
|
||||||
|
[phases.install]
|
||||||
|
cmds = ["npm install"]
|
||||||
|
|
||||||
|
[phases.build]
|
||||||
|
cmds = ["npm run build"]
|
||||||
|
|
||||||
|
[start]
|
||||||
|
cmd = "npm run preview"
|
||||||
2456
farsi_transcriber_web/package-lock.json
generated
Normal file
2456
farsi_transcriber_web/package-lock.json
generated
Normal file
File diff suppressed because it is too large
Load Diff
32
farsi_transcriber_web/package.json
Normal file
32
farsi_transcriber_web/package.json
Normal file
@ -0,0 +1,32 @@
|
|||||||
|
{
|
||||||
|
"name": "farsi-transcriber-web",
|
||||||
|
"private": true,
|
||||||
|
"version": "0.1.0",
|
||||||
|
"type": "module",
|
||||||
|
"scripts": {
|
||||||
|
"dev": "vite",
|
||||||
|
"build": "tsc -b && vite build",
|
||||||
|
"lint": "eslint . --ext ts,tsx --report-unused-disable-directives --max-warnings 0",
|
||||||
|
"preview": "vite preview"
|
||||||
|
},
|
||||||
|
"dependencies": {
|
||||||
|
"@tailwindcss/postcss": "^4.1.17",
|
||||||
|
"lucide-react": "^0.263.1",
|
||||||
|
"re-resizable": "^6.9.9",
|
||||||
|
"react": "^18.2.0",
|
||||||
|
"react-dom": "^18.2.0",
|
||||||
|
"sonner": "^1.2.0"
|
||||||
|
},
|
||||||
|
"devDependencies": {
|
||||||
|
"@types/node": "^20.8.0",
|
||||||
|
"@types/react": "^18.2.37",
|
||||||
|
"@types/react-dom": "^18.2.15",
|
||||||
|
"@vitejs/plugin-react": "^4.2.0",
|
||||||
|
"autoprefixer": "^10.4.16",
|
||||||
|
"postcss": "^8.4.31",
|
||||||
|
"tailwindcss": "^4.0.0",
|
||||||
|
"terser": "^5.44.1",
|
||||||
|
"typescript": "^5.2.2",
|
||||||
|
"vite": "^5.0.0"
|
||||||
|
}
|
||||||
|
}
|
||||||
6
farsi_transcriber_web/postcss.config.js
Normal file
6
farsi_transcriber_web/postcss.config.js
Normal file
@ -0,0 +1,6 @@
|
|||||||
|
export default {
|
||||||
|
plugins: {
|
||||||
|
'@tailwindcss/postcss': {},
|
||||||
|
autoprefixer: {},
|
||||||
|
},
|
||||||
|
}
|
||||||
13
farsi_transcriber_web/railway.toml
Normal file
13
farsi_transcriber_web/railway.toml
Normal file
@ -0,0 +1,13 @@
|
|||||||
|
# Railway configuration file
|
||||||
|
# https://docs.railway.app/reference/nixpacks
|
||||||
|
|
||||||
|
[build]
|
||||||
|
builder = "nixpacks"
|
||||||
|
|
||||||
|
[[services]]
|
||||||
|
name = "backend"
|
||||||
|
startCommand = "cd backend && gunicorn --workers 2 --worker-class sync --timeout 120 --bind 0.0.0.0:5000 app:app"
|
||||||
|
|
||||||
|
[[services]]
|
||||||
|
name = "frontend"
|
||||||
|
startCommand = "npm run build && npm run preview"
|
||||||
537
farsi_transcriber_web/src/App.tsx
Normal file
537
farsi_transcriber_web/src/App.tsx
Normal file
@ -0,0 +1,537 @@
|
|||||||
|
import { useState, useRef } from 'react';
|
||||||
|
import {
|
||||||
|
FileAudio,
|
||||||
|
Upload,
|
||||||
|
Moon,
|
||||||
|
Sun,
|
||||||
|
Search,
|
||||||
|
Copy,
|
||||||
|
X,
|
||||||
|
CheckCircle2,
|
||||||
|
Clock,
|
||||||
|
Loader2,
|
||||||
|
Download
|
||||||
|
} from 'lucide-react';
|
||||||
|
import { Resizable } from 're-resizable';
|
||||||
|
import { Toaster, toast } from 'sonner';
|
||||||
|
import Button from './components/Button';
|
||||||
|
import Progress from './components/Progress';
|
||||||
|
import Input from './components/Input';
|
||||||
|
import Select from './components/Select';
|
||||||
|
|
||||||
|
interface FileItem {
|
||||||
|
id: string;
|
||||||
|
name: string;
|
||||||
|
status: 'pending' | 'processing' | 'completed' | 'error';
|
||||||
|
progress?: number;
|
||||||
|
transcription?: TranscriptionSegment[];
|
||||||
|
file?: File;
|
||||||
|
fullText?: string;
|
||||||
|
}
|
||||||
|
|
||||||
|
interface TranscriptionSegment {
|
||||||
|
start: string;
|
||||||
|
end: string;
|
||||||
|
text: string;
|
||||||
|
}
|
||||||
|
|
||||||
|
// Get API URL from environment variable
|
||||||
|
const API_URL = import.meta.env.VITE_API_URL || 'http://localhost:5000';
|
||||||
|
|
||||||
|
export default function App() {
|
||||||
|
const [fileQueue, setFileQueue] = useState<FileItem[]>([]);
|
||||||
|
const [selectedFileId, setSelectedFileId] = useState<string | null>(null);
|
||||||
|
const [isDark, setIsDark] = useState(false);
|
||||||
|
const [windowSize, setWindowSize] = useState({ width: 1100, height: 700 });
|
||||||
|
const [searchQuery, setSearchQuery] = useState('');
|
||||||
|
const [exportFormat, setExportFormat] = useState('txt');
|
||||||
|
const fileInputRef = useRef<HTMLInputElement>(null);
|
||||||
|
|
||||||
|
// Theme colors
|
||||||
|
const theme = {
|
||||||
|
bg: isDark ? '#1a1a1a' : '#f5f5f5',
|
||||||
|
cardBg: isDark ? '#2d2d2d' : '#ffffff',
|
||||||
|
inputBg: isDark ? '#3a3a3a' : '#f9f9f9',
|
||||||
|
border: isDark ? '#4a4a4a' : '#d0d0d0',
|
||||||
|
text: isDark ? '#e0e0e0' : '#333333',
|
||||||
|
textSecondary: isDark ? '#a0a0a0' : '#666666',
|
||||||
|
progressBg: isDark ? '#404040' : '#e0e0e0',
|
||||||
|
sidebarBg: isDark ? '#252525' : '#fafafa',
|
||||||
|
hoverBg: isDark ? '#3a3a3a' : '#f0f0f0',
|
||||||
|
selectedBg: isDark ? '#4a4a4a' : '#e8f5e9',
|
||||||
|
};
|
||||||
|
|
||||||
|
const handleAddFiles = () => {
|
||||||
|
fileInputRef.current?.click();
|
||||||
|
};
|
||||||
|
|
||||||
|
const handleFileInputChange = (e: React.ChangeEvent<HTMLInputElement>) => {
|
||||||
|
const files = e.currentTarget.files;
|
||||||
|
if (!files) return;
|
||||||
|
|
||||||
|
const newFiles: FileItem[] = [];
|
||||||
|
for (let i = 0; i < files.length; i++) {
|
||||||
|
const file = files[i];
|
||||||
|
const newFileItem: FileItem = {
|
||||||
|
id: `${Date.now()}-${i}`,
|
||||||
|
name: file.name,
|
||||||
|
status: 'pending',
|
||||||
|
file: file,
|
||||||
|
};
|
||||||
|
newFiles.push(newFileItem);
|
||||||
|
}
|
||||||
|
|
||||||
|
setFileQueue([...fileQueue, ...newFiles]);
|
||||||
|
if (!selectedFileId && newFiles.length > 0) {
|
||||||
|
setSelectedFileId(newFiles[0].id);
|
||||||
|
}
|
||||||
|
toast.success(`${newFiles.length} file(s) added to queue`);
|
||||||
|
|
||||||
|
// Reset input
|
||||||
|
if (fileInputRef.current) {
|
||||||
|
fileInputRef.current.value = '';
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
const handleRemoveFile = (id: string) => {
|
||||||
|
setFileQueue(fileQueue.filter(f => f.id !== id));
|
||||||
|
if (selectedFileId === id) {
|
||||||
|
setSelectedFileId(fileQueue[0]?.id || null);
|
||||||
|
}
|
||||||
|
toast.info('File removed from queue');
|
||||||
|
};
|
||||||
|
|
||||||
|
const handleTranscribe = async () => {
|
||||||
|
if (!selectedFileId) return;
|
||||||
|
|
||||||
|
const fileIndex = fileQueue.findIndex(f => f.id === selectedFileId);
|
||||||
|
if (fileIndex === -1 || !fileQueue[fileIndex].file) return;
|
||||||
|
|
||||||
|
// Update status to processing
|
||||||
|
const updatedQueue = [...fileQueue];
|
||||||
|
updatedQueue[fileIndex].status = 'processing';
|
||||||
|
updatedQueue[fileIndex].progress = 0;
|
||||||
|
setFileQueue(updatedQueue);
|
||||||
|
|
||||||
|
try {
|
||||||
|
const file = fileQueue[fileIndex].file!;
|
||||||
|
const formData = new FormData();
|
||||||
|
formData.append('file', file);
|
||||||
|
formData.append('language', 'fa'); // Farsi by default
|
||||||
|
|
||||||
|
// Show loading toast
|
||||||
|
const loadingToastId = toast.loading('Loading Whisper model (first time only)...');
|
||||||
|
|
||||||
|
const response = await fetch(`${API_URL}/transcribe`, {
|
||||||
|
method: 'POST',
|
||||||
|
body: formData,
|
||||||
|
});
|
||||||
|
|
||||||
|
if (!response.ok) {
|
||||||
|
throw new Error(`API error: ${response.statusText}`);
|
||||||
|
}
|
||||||
|
|
||||||
|
const result = await response.json();
|
||||||
|
|
||||||
|
// Dismiss loading toast
|
||||||
|
toast.dismiss(loadingToastId);
|
||||||
|
|
||||||
|
if (result.status === 'success') {
|
||||||
|
const updated = [...fileQueue];
|
||||||
|
updated[fileIndex].status = 'completed';
|
||||||
|
updated[fileIndex].progress = 100;
|
||||||
|
updated[fileIndex].transcription = result.segments;
|
||||||
|
updated[fileIndex].fullText = result.text;
|
||||||
|
setFileQueue(updated);
|
||||||
|
toast.success('Transcription completed!');
|
||||||
|
} else {
|
||||||
|
throw new Error(result.error || 'Unknown error');
|
||||||
|
}
|
||||||
|
} catch (error) {
|
||||||
|
const updated = [...fileQueue];
|
||||||
|
updated[fileIndex].status = 'error';
|
||||||
|
setFileQueue(updated);
|
||||||
|
const errorMsg = error instanceof Error ? error.message : 'Failed to transcribe file';
|
||||||
|
toast.error(errorMsg);
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
const handleCopySegment = (text: string) => {
|
||||||
|
navigator.clipboard.writeText(text);
|
||||||
|
toast.success('Copied to clipboard');
|
||||||
|
};
|
||||||
|
|
||||||
|
const handleExport = async () => {
|
||||||
|
const selectedFile = fileQueue.find(f => f.id === selectedFileId);
|
||||||
|
if (!selectedFile?.transcription || !selectedFile.fullText) {
|
||||||
|
toast.error('No transcription to export');
|
||||||
|
return;
|
||||||
|
}
|
||||||
|
|
||||||
|
try {
|
||||||
|
const toastId = toast.loading('Preparing export...');
|
||||||
|
|
||||||
|
const response = await fetch(`${API_URL}/export`, {
|
||||||
|
method: 'POST',
|
||||||
|
headers: {
|
||||||
|
'Content-Type': 'application/json',
|
||||||
|
},
|
||||||
|
body: JSON.stringify({
|
||||||
|
transcription: selectedFile.fullText,
|
||||||
|
segments: selectedFile.transcription,
|
||||||
|
format: exportFormat,
|
||||||
|
}),
|
||||||
|
});
|
||||||
|
|
||||||
|
if (!response.ok) {
|
||||||
|
throw new Error(`Export failed: ${response.statusText}`);
|
||||||
|
}
|
||||||
|
|
||||||
|
const result = await response.json();
|
||||||
|
toast.dismiss(toastId);
|
||||||
|
|
||||||
|
if (result.status === 'success') {
|
||||||
|
// Create a blob and download
|
||||||
|
const blob = new Blob([result.content], { type: result.mime_type });
|
||||||
|
const url = window.URL.createObjectURL(blob);
|
||||||
|
const a = document.createElement('a');
|
||||||
|
a.href = url;
|
||||||
|
a.download = `${selectedFile.name.split('.')[0]}.${exportFormat === 'json' ? 'json' : exportFormat}`;
|
||||||
|
document.body.appendChild(a);
|
||||||
|
a.click();
|
||||||
|
window.URL.revokeObjectURL(url);
|
||||||
|
document.body.removeChild(a);
|
||||||
|
toast.success(`Exported as ${exportFormat.toUpperCase()}`);
|
||||||
|
} else {
|
||||||
|
throw new Error(result.error || 'Export failed');
|
||||||
|
}
|
||||||
|
} catch (error) {
|
||||||
|
const errorMsg = error instanceof Error ? error.message : 'Failed to export';
|
||||||
|
toast.error(errorMsg);
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
const handleClearAll = () => {
|
||||||
|
setFileQueue([]);
|
||||||
|
setSelectedFileId(null);
|
||||||
|
setSearchQuery('');
|
||||||
|
toast.info('All files cleared');
|
||||||
|
};
|
||||||
|
|
||||||
|
const selectedFile = fileQueue.find(f => f.id === selectedFileId);
|
||||||
|
const currentTranscription = selectedFile?.transcription || [];
|
||||||
|
|
||||||
|
// Filter transcription based on search
|
||||||
|
const filteredTranscription = searchQuery
|
||||||
|
? currentTranscription.filter(seg =>
|
||||||
|
seg.text.toLowerCase().includes(searchQuery.toLowerCase())
|
||||||
|
)
|
||||||
|
: currentTranscription;
|
||||||
|
|
||||||
|
// Function to highlight search text
|
||||||
|
const highlightText = (text: string, query: string) => {
|
||||||
|
if (!query) return text;
|
||||||
|
|
||||||
|
const parts = text.split(new RegExp(`(${query})`, 'gi'));
|
||||||
|
return parts.map((part) =>
|
||||||
|
part.toLowerCase() === query.toLowerCase()
|
||||||
|
? `<mark style="background-color: ${isDark ? '#4CAF50' : '#FFEB3B'}; color: ${isDark ? '#000' : '#000'}; padding: 2px 4px; border-radius: 2px;">${part}</mark>`
|
||||||
|
: part
|
||||||
|
).join('');
|
||||||
|
};
|
||||||
|
|
||||||
|
const getStatusIcon = (status: FileItem['status']) => {
|
||||||
|
switch (status) {
|
||||||
|
case 'completed':
|
||||||
|
return <CheckCircle2 className="w-4 h-4 text-green-500" />;
|
||||||
|
case 'processing':
|
||||||
|
return <Loader2 className="w-4 h-4 text-blue-500 animate-spin" />;
|
||||||
|
case 'error':
|
||||||
|
return <X className="w-4 h-4 text-red-500" />;
|
||||||
|
default:
|
||||||
|
return <Clock className="w-4 h-4" style={{ color: theme.textSecondary }} />;
|
||||||
|
}
|
||||||
|
};
|
||||||
|
|
||||||
|
return (
|
||||||
|
<div className="min-h-screen flex items-center justify-center p-8" style={{ backgroundColor: theme.bg }}>
|
||||||
|
<Toaster theme={isDark ? 'dark' : 'light'} position="top-right" />
|
||||||
|
|
||||||
|
{/* Hidden file input */}
|
||||||
|
<input
|
||||||
|
ref={fileInputRef}
|
||||||
|
type="file"
|
||||||
|
multiple
|
||||||
|
accept="audio/*,video/*"
|
||||||
|
onChange={handleFileInputChange}
|
||||||
|
style={{ display: 'none' }}
|
||||||
|
/>
|
||||||
|
|
||||||
|
<Resizable
|
||||||
|
size={windowSize}
|
||||||
|
onResizeStop={(_e, _direction, _ref, d) => {
|
||||||
|
setWindowSize({
|
||||||
|
width: windowSize.width + d.width,
|
||||||
|
height: windowSize.height + d.height,
|
||||||
|
});
|
||||||
|
}}
|
||||||
|
minWidth={900}
|
||||||
|
minHeight={600}
|
||||||
|
className="rounded-lg shadow-2xl overflow-hidden"
|
||||||
|
style={{
|
||||||
|
backgroundColor: theme.cardBg,
|
||||||
|
border: `2px solid ${theme.border}`,
|
||||||
|
}}
|
||||||
|
handleStyles={{
|
||||||
|
right: { cursor: 'ew-resize' },
|
||||||
|
bottom: { cursor: 'ns-resize' },
|
||||||
|
bottomRight: { cursor: 'nwse-resize' },
|
||||||
|
}}
|
||||||
|
>
|
||||||
|
<div className="flex h-full">
|
||||||
|
{/* Left Sidebar - File Queue */}
|
||||||
|
<div
|
||||||
|
className="w-64 border-r flex flex-col overflow-hidden"
|
||||||
|
style={{ borderColor: theme.border, backgroundColor: theme.sidebarBg }}
|
||||||
|
>
|
||||||
|
<div className="p-4 border-b" style={{ borderColor: theme.border }}>
|
||||||
|
<h3 className="mb-3 font-semibold" style={{ color: theme.text }}>
|
||||||
|
File Queue
|
||||||
|
</h3>
|
||||||
|
<Button
|
||||||
|
onClick={handleAddFiles}
|
||||||
|
className="w-full bg-green-500 hover:bg-green-600 text-white"
|
||||||
|
>
|
||||||
|
<Upload className="w-4 h-4 mr-2" />
|
||||||
|
Add Files
|
||||||
|
</Button>
|
||||||
|
</div>
|
||||||
|
|
||||||
|
<div className="flex-1 overflow-auto p-2">
|
||||||
|
{fileQueue.length === 0 ? (
|
||||||
|
<p className="text-center text-xs p-4" style={{ color: theme.textSecondary }}>
|
||||||
|
No files in queue
|
||||||
|
</p>
|
||||||
|
) : (
|
||||||
|
fileQueue.map((file) => (
|
||||||
|
<div
|
||||||
|
key={file.id}
|
||||||
|
className="mb-2 p-3 rounded-lg cursor-pointer transition-colors border"
|
||||||
|
style={{
|
||||||
|
backgroundColor: selectedFileId === file.id ? theme.selectedBg : theme.cardBg,
|
||||||
|
borderColor: selectedFileId === file.id ? '#4CAF50' : theme.border,
|
||||||
|
}}
|
||||||
|
onClick={() => setSelectedFileId(file.id)}
|
||||||
|
>
|
||||||
|
<div className="flex items-start justify-between gap-2 mb-2">
|
||||||
|
<div className="flex items-center gap-2 flex-1 min-w-0">
|
||||||
|
{getStatusIcon(file.status)}
|
||||||
|
<span className="text-xs truncate" style={{ color: theme.text }}>
|
||||||
|
{file.name}
|
||||||
|
</span>
|
||||||
|
</div>
|
||||||
|
<button
|
||||||
|
onClick={(e) => {
|
||||||
|
e.stopPropagation();
|
||||||
|
handleRemoveFile(file.id);
|
||||||
|
}}
|
||||||
|
className="hover:opacity-70"
|
||||||
|
>
|
||||||
|
<X className="w-3 h-3" style={{ color: theme.textSecondary }} />
|
||||||
|
</button>
|
||||||
|
</div>
|
||||||
|
{file.status === 'processing' && (
|
||||||
|
<div className="space-y-1">
|
||||||
|
<Progress value={file.progress || 0} />
|
||||||
|
<p className="text-xs" style={{ color: theme.textSecondary }}>
|
||||||
|
{file.progress}%
|
||||||
|
</p>
|
||||||
|
</div>
|
||||||
|
)}
|
||||||
|
</div>
|
||||||
|
))
|
||||||
|
)}
|
||||||
|
</div>
|
||||||
|
</div>
|
||||||
|
|
||||||
|
{/* Main Content Area */}
|
||||||
|
<div className="flex-1 flex flex-col overflow-hidden">
|
||||||
|
{/* Header */}
|
||||||
|
<div
|
||||||
|
className="p-5 border-b flex items-center justify-between"
|
||||||
|
style={{ borderColor: theme.border }}
|
||||||
|
>
|
||||||
|
<div className="flex items-center gap-3">
|
||||||
|
<h1 style={{ color: theme.text }} className="text-lg font-semibold">
|
||||||
|
Farsi Audio/Video Transcriber
|
||||||
|
</h1>
|
||||||
|
<span className="text-xs" style={{ color: theme.textSecondary }}>
|
||||||
|
{windowSize.width}×{windowSize.height}
|
||||||
|
</span>
|
||||||
|
</div>
|
||||||
|
<Button
|
||||||
|
onClick={() => setIsDark(!isDark)}
|
||||||
|
variant="outline"
|
||||||
|
style={{ borderColor: theme.border, backgroundColor: theme.cardBg }}
|
||||||
|
>
|
||||||
|
{isDark ? (
|
||||||
|
<Sun className="w-4 h-4" style={{ color: theme.text }} />
|
||||||
|
) : (
|
||||||
|
<Moon className="w-4 h-4" style={{ color: theme.text }} />
|
||||||
|
)}
|
||||||
|
</Button>
|
||||||
|
</div>
|
||||||
|
|
||||||
|
<div className="flex-1 flex flex-col p-5 overflow-hidden">
|
||||||
|
{/* File Info & Actions */}
|
||||||
|
<div
|
||||||
|
className="mb-4 p-4 rounded-lg border"
|
||||||
|
style={{ backgroundColor: theme.inputBg, borderColor: theme.border }}
|
||||||
|
>
|
||||||
|
<div className="flex items-center justify-between">
|
||||||
|
<div className="flex items-center gap-3">
|
||||||
|
<FileAudio className="w-5 h-5" style={{ color: theme.textSecondary }} />
|
||||||
|
<div>
|
||||||
|
<p className="text-sm" style={{ color: theme.text }}>
|
||||||
|
{selectedFile ? selectedFile.name : 'No file selected'}
|
||||||
|
</p>
|
||||||
|
{selectedFile?.status === 'processing' && (
|
||||||
|
<p className="text-xs" style={{ color: theme.textSecondary }}>
|
||||||
|
Processing... {selectedFile.progress}%
|
||||||
|
</p>
|
||||||
|
)}
|
||||||
|
{selectedFile?.status === 'completed' && (
|
||||||
|
<p className="text-xs text-green-500">Completed</p>
|
||||||
|
)}
|
||||||
|
</div>
|
||||||
|
</div>
|
||||||
|
<Button
|
||||||
|
onClick={handleTranscribe}
|
||||||
|
disabled={!selectedFile || selectedFile.status === 'processing' || selectedFile.status === 'completed'}
|
||||||
|
className="bg-green-500 hover:bg-green-600 text-white disabled:bg-gray-400 disabled:cursor-not-allowed"
|
||||||
|
>
|
||||||
|
{selectedFile?.status === 'processing' ? 'Transcribing...' : 'Transcribe'}
|
||||||
|
</Button>
|
||||||
|
</div>
|
||||||
|
</div>
|
||||||
|
|
||||||
|
{/* Search & Export Controls */}
|
||||||
|
{selectedFile?.transcription && (
|
||||||
|
<div className="mb-4 flex gap-2">
|
||||||
|
<div className="flex-1 relative">
|
||||||
|
<Search
|
||||||
|
className="w-4 h-4 absolute left-3 top-1/2 -translate-y-1/2"
|
||||||
|
style={{ color: theme.textSecondary }}
|
||||||
|
/>
|
||||||
|
<Input
|
||||||
|
placeholder="Search in transcription..."
|
||||||
|
value={searchQuery}
|
||||||
|
onChange={(e) => setSearchQuery(e.target.value)}
|
||||||
|
style={{
|
||||||
|
backgroundColor: theme.inputBg,
|
||||||
|
borderColor: theme.border,
|
||||||
|
color: theme.text,
|
||||||
|
paddingLeft: '2.25rem',
|
||||||
|
}}
|
||||||
|
/>
|
||||||
|
</div>
|
||||||
|
<Select
|
||||||
|
value={exportFormat}
|
||||||
|
onChange={(e) => setExportFormat(e.target.value as 'txt' | 'srt' | 'vtt' | 'json')}
|
||||||
|
>
|
||||||
|
<option value="txt">TXT</option>
|
||||||
|
<option value="srt">SRT</option>
|
||||||
|
<option value="vtt">VTT</option>
|
||||||
|
<option value="json">JSON</option>
|
||||||
|
</Select>
|
||||||
|
<Button
|
||||||
|
onClick={handleExport}
|
||||||
|
variant="outline"
|
||||||
|
style={{ borderColor: theme.border, backgroundColor: theme.cardBg, color: theme.text }}
|
||||||
|
>
|
||||||
|
<Download className="w-4 h-4 mr-2" />
|
||||||
|
Export
|
||||||
|
</Button>
|
||||||
|
</div>
|
||||||
|
)}
|
||||||
|
|
||||||
|
{/* Transcription Results */}
|
||||||
|
<div className="flex-1 flex flex-col min-h-0">
|
||||||
|
<div className="flex items-center justify-between mb-2">
|
||||||
|
<label style={{ color: theme.text }} className="text-sm font-medium">
|
||||||
|
Transcription Results:
|
||||||
|
</label>
|
||||||
|
{searchQuery && (
|
||||||
|
<span className="text-xs" style={{ color: theme.textSecondary }}>
|
||||||
|
{filteredTranscription.length} results found
|
||||||
|
</span>
|
||||||
|
)}
|
||||||
|
</div>
|
||||||
|
|
||||||
|
<div
|
||||||
|
className="flex-1 rounded-lg border p-4 overflow-auto"
|
||||||
|
style={{ backgroundColor: theme.cardBg, borderColor: theme.border }}
|
||||||
|
>
|
||||||
|
{currentTranscription.length === 0 ? (
|
||||||
|
<p className="text-center" style={{ color: theme.textSecondary }}>
|
||||||
|
Transcription results will appear here...
|
||||||
|
</p>
|
||||||
|
) : (
|
||||||
|
<div className="space-y-3">
|
||||||
|
{filteredTranscription.map((segment, index) => (
|
||||||
|
<div
|
||||||
|
key={index}
|
||||||
|
className="p-3 rounded-md border group hover:shadow-sm transition-shadow"
|
||||||
|
style={{
|
||||||
|
backgroundColor: theme.inputBg,
|
||||||
|
borderColor: theme.border,
|
||||||
|
}}
|
||||||
|
>
|
||||||
|
<div className="flex items-start justify-between gap-3 mb-2">
|
||||||
|
<span
|
||||||
|
className="text-xs font-mono"
|
||||||
|
style={{ color: theme.textSecondary }}
|
||||||
|
>
|
||||||
|
[{segment.start} - {segment.end}]
|
||||||
|
</span>
|
||||||
|
<button
|
||||||
|
onClick={() => handleCopySegment(segment.text)}
|
||||||
|
className="opacity-0 group-hover:opacity-100 transition-opacity"
|
||||||
|
title="Copy segment"
|
||||||
|
>
|
||||||
|
<Copy className="w-3 h-3" style={{ color: theme.textSecondary }} />
|
||||||
|
</button>
|
||||||
|
</div>
|
||||||
|
<p
|
||||||
|
className="text-sm leading-relaxed"
|
||||||
|
style={{ color: theme.text }}
|
||||||
|
dir="rtl"
|
||||||
|
dangerouslySetInnerHTML={{ __html: highlightText(segment.text, searchQuery) }}
|
||||||
|
/>
|
||||||
|
</div>
|
||||||
|
))}
|
||||||
|
</div>
|
||||||
|
)}
|
||||||
|
</div>
|
||||||
|
</div>
|
||||||
|
|
||||||
|
{/* Bottom Actions */}
|
||||||
|
<div className="flex justify-between items-center mt-4">
|
||||||
|
<p className="text-xs" style={{ color: theme.textSecondary }}>
|
||||||
|
{selectedFile?.status === 'completed' && `${currentTranscription.length} segments`}
|
||||||
|
</p>
|
||||||
|
<Button
|
||||||
|
onClick={handleClearAll}
|
||||||
|
variant="outline"
|
||||||
|
style={{ borderColor: theme.border, backgroundColor: theme.cardBg, color: theme.text }}
|
||||||
|
>
|
||||||
|
Clear All
|
||||||
|
</Button>
|
||||||
|
</div>
|
||||||
|
</div>
|
||||||
|
</div>
|
||||||
|
</div>
|
||||||
|
</Resizable>
|
||||||
|
</div>
|
||||||
|
);
|
||||||
|
}
|
||||||
36
farsi_transcriber_web/src/components/Button.tsx
Normal file
36
farsi_transcriber_web/src/components/Button.tsx
Normal file
@ -0,0 +1,36 @@
|
|||||||
|
import React from 'react';
|
||||||
|
|
||||||
|
interface ButtonProps extends React.ButtonHTMLAttributes<HTMLButtonElement> {
|
||||||
|
variant?: 'default' | 'outline';
|
||||||
|
size?: 'sm' | 'md' | 'lg';
|
||||||
|
children: React.ReactNode;
|
||||||
|
}
|
||||||
|
|
||||||
|
const Button = React.forwardRef<HTMLButtonElement, ButtonProps>(
|
||||||
|
({ variant = 'default', size = 'md', className, ...props }, ref) => {
|
||||||
|
const baseStyles = 'font-medium rounded transition-colors disabled:opacity-50 disabled:cursor-not-allowed inline-flex items-center justify-center';
|
||||||
|
|
||||||
|
const variantStyles = {
|
||||||
|
default: 'bg-green-500 hover:bg-green-600 text-white',
|
||||||
|
outline: 'border border-gray-300 hover:bg-gray-100 text-gray-900',
|
||||||
|
};
|
||||||
|
|
||||||
|
const sizeStyles = {
|
||||||
|
sm: 'px-3 py-1.5 text-sm',
|
||||||
|
md: 'px-4 py-2 text-base',
|
||||||
|
lg: 'px-6 py-3 text-lg',
|
||||||
|
};
|
||||||
|
|
||||||
|
return (
|
||||||
|
<button
|
||||||
|
ref={ref}
|
||||||
|
className={`${baseStyles} ${variantStyles[variant]} ${sizeStyles[size]} ${className || ''}`}
|
||||||
|
{...props}
|
||||||
|
/>
|
||||||
|
);
|
||||||
|
}
|
||||||
|
);
|
||||||
|
|
||||||
|
Button.displayName = 'Button';
|
||||||
|
|
||||||
|
export default Button;
|
||||||
24
farsi_transcriber_web/src/components/Input.tsx
Normal file
24
farsi_transcriber_web/src/components/Input.tsx
Normal file
@ -0,0 +1,24 @@
|
|||||||
|
import React from 'react';
|
||||||
|
|
||||||
|
interface InputProps extends React.InputHTMLAttributes<HTMLInputElement> {
|
||||||
|
label?: string;
|
||||||
|
}
|
||||||
|
|
||||||
|
const Input = React.forwardRef<HTMLInputElement, InputProps>(
|
||||||
|
({ label, className, ...props }, ref) => {
|
||||||
|
return (
|
||||||
|
<div className="w-full">
|
||||||
|
{label && <label className="block text-sm font-medium mb-1">{label}</label>}
|
||||||
|
<input
|
||||||
|
ref={ref}
|
||||||
|
className={`w-full px-3 py-2 border rounded-md text-sm focus:outline-none focus:ring-2 focus:ring-green-500 ${className || ''}`}
|
||||||
|
{...props}
|
||||||
|
/>
|
||||||
|
</div>
|
||||||
|
);
|
||||||
|
}
|
||||||
|
);
|
||||||
|
|
||||||
|
Input.displayName = 'Input';
|
||||||
|
|
||||||
|
export default Input;
|
||||||
15
farsi_transcriber_web/src/components/Progress.tsx
Normal file
15
farsi_transcriber_web/src/components/Progress.tsx
Normal file
@ -0,0 +1,15 @@
|
|||||||
|
interface ProgressProps {
|
||||||
|
value: number;
|
||||||
|
className?: string;
|
||||||
|
}
|
||||||
|
|
||||||
|
export default function Progress({ value, className }: ProgressProps) {
|
||||||
|
return (
|
||||||
|
<div className={`w-full bg-gray-200 rounded-full h-1.5 overflow-hidden ${className || ''}`}>
|
||||||
|
<div
|
||||||
|
className="bg-green-500 h-full transition-all duration-300"
|
||||||
|
style={{ width: `${Math.min(100, Math.max(0, value))}%` }}
|
||||||
|
/>
|
||||||
|
</div>
|
||||||
|
);
|
||||||
|
}
|
||||||
27
farsi_transcriber_web/src/components/Select.tsx
Normal file
27
farsi_transcriber_web/src/components/Select.tsx
Normal file
@ -0,0 +1,27 @@
|
|||||||
|
import React from 'react';
|
||||||
|
|
||||||
|
interface SelectProps extends React.SelectHTMLAttributes<HTMLSelectElement> {
|
||||||
|
label?: string;
|
||||||
|
children: React.ReactNode;
|
||||||
|
}
|
||||||
|
|
||||||
|
const Select = React.forwardRef<HTMLSelectElement, SelectProps>(
|
||||||
|
({ label, className, children, ...props }, ref) => {
|
||||||
|
return (
|
||||||
|
<div className="w-full">
|
||||||
|
{label && <label className="block text-sm font-medium mb-1">{label}</label>}
|
||||||
|
<select
|
||||||
|
ref={ref}
|
||||||
|
className={`w-full px-3 py-2 border rounded-md text-sm focus:outline-none focus:ring-2 focus:ring-green-500 bg-white ${className || ''}`}
|
||||||
|
{...props}
|
||||||
|
>
|
||||||
|
{children}
|
||||||
|
</select>
|
||||||
|
</div>
|
||||||
|
);
|
||||||
|
}
|
||||||
|
);
|
||||||
|
|
||||||
|
Select.displayName = 'Select';
|
||||||
|
|
||||||
|
export default Select;
|
||||||
4
farsi_transcriber_web/src/components/__init__.ts
Normal file
4
farsi_transcriber_web/src/components/__init__.ts
Normal file
@ -0,0 +1,4 @@
|
|||||||
|
export { default as Button } from './Button';
|
||||||
|
export { default as Progress } from './Progress';
|
||||||
|
export { default as Input } from './Input';
|
||||||
|
export { default as Select } from './Select';
|
||||||
46
farsi_transcriber_web/src/index.css
Normal file
46
farsi_transcriber_web/src/index.css
Normal file
@ -0,0 +1,46 @@
|
|||||||
|
@import "tailwindcss";
|
||||||
|
|
||||||
|
* {
|
||||||
|
margin: 0;
|
||||||
|
padding: 0;
|
||||||
|
box-sizing: border-box;
|
||||||
|
}
|
||||||
|
|
||||||
|
html, body, #root {
|
||||||
|
width: 100%;
|
||||||
|
height: 100%;
|
||||||
|
font-family: -apple-system, BlinkMacSystemFont, 'Segoe UI', 'Roboto', 'Oxygen',
|
||||||
|
'Ubuntu', 'Cantarell', 'Fira Sans', 'Droid Sans', 'Helvetica Neue',
|
||||||
|
sans-serif;
|
||||||
|
-webkit-font-smoothing: antialiased;
|
||||||
|
-moz-osx-font-smoothing: grayscale;
|
||||||
|
}
|
||||||
|
|
||||||
|
mark {
|
||||||
|
display: inline-block;
|
||||||
|
}
|
||||||
|
|
||||||
|
/* RTL Support */
|
||||||
|
[dir="rtl"] {
|
||||||
|
text-align: right;
|
||||||
|
direction: rtl;
|
||||||
|
}
|
||||||
|
|
||||||
|
/* Scrollbar styling */
|
||||||
|
::-webkit-scrollbar {
|
||||||
|
width: 8px;
|
||||||
|
height: 8px;
|
||||||
|
}
|
||||||
|
|
||||||
|
::-webkit-scrollbar-track {
|
||||||
|
background: transparent;
|
||||||
|
}
|
||||||
|
|
||||||
|
::-webkit-scrollbar-thumb {
|
||||||
|
background: #4CAF50;
|
||||||
|
border-radius: 4px;
|
||||||
|
}
|
||||||
|
|
||||||
|
::-webkit-scrollbar-thumb:hover {
|
||||||
|
background: #45a049;
|
||||||
|
}
|
||||||
10
farsi_transcriber_web/src/main.tsx
Normal file
10
farsi_transcriber_web/src/main.tsx
Normal file
@ -0,0 +1,10 @@
|
|||||||
|
import React from 'react'
|
||||||
|
import ReactDOM from 'react-dom/client'
|
||||||
|
import App from './App.tsx'
|
||||||
|
import './index.css'
|
||||||
|
|
||||||
|
ReactDOM.createRoot(document.getElementById('root')!).render(
|
||||||
|
<React.StrictMode>
|
||||||
|
<App />
|
||||||
|
</React.StrictMode>,
|
||||||
|
)
|
||||||
17
farsi_transcriber_web/tailwind.config.js
Normal file
17
farsi_transcriber_web/tailwind.config.js
Normal file
@ -0,0 +1,17 @@
|
|||||||
|
/** @type {import('tailwindcss').Config} */
|
||||||
|
export default {
|
||||||
|
content: [
|
||||||
|
"./index.html",
|
||||||
|
"./src/**/*.{js,ts,jsx,tsx}",
|
||||||
|
],
|
||||||
|
theme: {
|
||||||
|
extend: {
|
||||||
|
colors: {
|
||||||
|
primary: '#4CAF50',
|
||||||
|
'primary-hover': '#45a049',
|
||||||
|
'primary-active': '#3d8b40',
|
||||||
|
}
|
||||||
|
},
|
||||||
|
},
|
||||||
|
plugins: [],
|
||||||
|
}
|
||||||
27
farsi_transcriber_web/tsconfig.json
Normal file
27
farsi_transcriber_web/tsconfig.json
Normal file
@ -0,0 +1,27 @@
|
|||||||
|
{
|
||||||
|
"compilerOptions": {
|
||||||
|
"target": "ES2020",
|
||||||
|
"useDefineForClassFields": true,
|
||||||
|
"lib": ["ES2020", "DOM", "DOM.Iterable"],
|
||||||
|
"module": "ESNext",
|
||||||
|
"skipLibCheck": true,
|
||||||
|
"esModuleInterop": true,
|
||||||
|
"allowSyntheticDefaultImports": true,
|
||||||
|
|
||||||
|
/* Bundler mode */
|
||||||
|
"moduleResolution": "bundler",
|
||||||
|
"allowImportingTsExtensions": true,
|
||||||
|
"resolveJsonModule": true,
|
||||||
|
"isolatedModules": true,
|
||||||
|
"noEmit": true,
|
||||||
|
"jsx": "react-jsx",
|
||||||
|
|
||||||
|
/* Linting */
|
||||||
|
"strict": true,
|
||||||
|
"noUnusedLocals": true,
|
||||||
|
"noUnusedParameters": true,
|
||||||
|
"noFallthroughCasesInSwitch": true
|
||||||
|
},
|
||||||
|
"include": ["src"],
|
||||||
|
"references": [{ "path": "./tsconfig.node.json" }]
|
||||||
|
}
|
||||||
10
farsi_transcriber_web/tsconfig.node.json
Normal file
10
farsi_transcriber_web/tsconfig.node.json
Normal file
@ -0,0 +1,10 @@
|
|||||||
|
{
|
||||||
|
"compilerOptions": {
|
||||||
|
"composite": true,
|
||||||
|
"skipLibCheck": true,
|
||||||
|
"module": "ESNext",
|
||||||
|
"moduleResolution": "bundler",
|
||||||
|
"allowSyntheticDefaultImports": true
|
||||||
|
},
|
||||||
|
"include": ["vite.config.ts"]
|
||||||
|
}
|
||||||
38
farsi_transcriber_web/vite.config.ts
Normal file
38
farsi_transcriber_web/vite.config.ts
Normal file
@ -0,0 +1,38 @@
|
|||||||
|
import { defineConfig } from 'vite'
|
||||||
|
import react from '@vitejs/plugin-react'
|
||||||
|
|
||||||
|
// https://vitejs.dev/config/
|
||||||
|
export default defineConfig({
|
||||||
|
plugins: [react()],
|
||||||
|
server: {
|
||||||
|
port: 3000,
|
||||||
|
proxy: {
|
||||||
|
'/api': {
|
||||||
|
target: 'http://localhost:5000',
|
||||||
|
changeOrigin: true,
|
||||||
|
rewrite: (path) => path.replace(/^\/api/, '')
|
||||||
|
}
|
||||||
|
}
|
||||||
|
},
|
||||||
|
build: {
|
||||||
|
outDir: 'dist',
|
||||||
|
sourcemap: false,
|
||||||
|
minify: 'terser',
|
||||||
|
terserOptions: {
|
||||||
|
compress: {
|
||||||
|
drop_console: true,
|
||||||
|
},
|
||||||
|
},
|
||||||
|
rollupOptions: {
|
||||||
|
output: {
|
||||||
|
manualChunks: {
|
||||||
|
vendor: ['react', 'react-dom'],
|
||||||
|
},
|
||||||
|
},
|
||||||
|
},
|
||||||
|
},
|
||||||
|
preview: {
|
||||||
|
port: parseInt(process.env.PORT || '3000'),
|
||||||
|
host: '0.0.0.0',
|
||||||
|
}
|
||||||
|
})
|
||||||
@ -281,7 +281,7 @@ def transcribe(
|
|||||||
time_offset = float(seek * HOP_LENGTH / SAMPLE_RATE)
|
time_offset = float(seek * HOP_LENGTH / SAMPLE_RATE)
|
||||||
window_end_time = float((seek + N_FRAMES) * HOP_LENGTH / SAMPLE_RATE)
|
window_end_time = float((seek + N_FRAMES) * HOP_LENGTH / SAMPLE_RATE)
|
||||||
segment_size = min(N_FRAMES, content_frames - seek, seek_clip_end - seek)
|
segment_size = min(N_FRAMES, content_frames - seek, seek_clip_end - seek)
|
||||||
mel_segment = mel[:, seek : seek + segment_size]
|
mel_segment = mel[:, seek:seek + segment_size]
|
||||||
segment_duration = segment_size * HOP_LENGTH / SAMPLE_RATE
|
segment_duration = segment_size * HOP_LENGTH / SAMPLE_RATE
|
||||||
mel_segment = pad_or_trim(mel_segment, N_FRAMES).to(model.device).to(dtype)
|
mel_segment = pad_or_trim(mel_segment, N_FRAMES).to(model.device).to(dtype)
|
||||||
|
|
||||||
@ -444,7 +444,7 @@ def transcribe(
|
|||||||
continue
|
continue
|
||||||
if is_segment_anomaly(segment):
|
if is_segment_anomaly(segment):
|
||||||
next_segment = next_words_segment(
|
next_segment = next_words_segment(
|
||||||
current_segments[si + 1 :]
|
current_segments[si + 1:]
|
||||||
)
|
)
|
||||||
if next_segment is not None:
|
if next_segment is not None:
|
||||||
hal_next_start = next_segment["words"][0]["start"]
|
hal_next_start = next_segment["words"][0]["start"]
|
||||||
@ -508,7 +508,7 @@ def transcribe(
|
|||||||
pbar.update(min(content_frames, seek) - previous_seek)
|
pbar.update(min(content_frames, seek) - previous_seek)
|
||||||
|
|
||||||
return dict(
|
return dict(
|
||||||
text=tokenizer.decode(all_tokens[len(initial_prompt_tokens) :]),
|
text=tokenizer.decode(all_tokens[len(initial_prompt_tokens):]),
|
||||||
segments=all_segments,
|
segments=all_segments,
|
||||||
language=language,
|
language=language,
|
||||||
)
|
)
|
||||||
|
|||||||
@ -153,7 +153,7 @@ class SubtitlesWriter(ResultWriter):
|
|||||||
if max_words_per_line > len(segment["words"]) - chunk_index:
|
if max_words_per_line > len(segment["words"]) - chunk_index:
|
||||||
words_count = remaining_words
|
words_count = remaining_words
|
||||||
for i, original_timing in enumerate(
|
for i, original_timing in enumerate(
|
||||||
segment["words"][chunk_index : chunk_index + words_count]
|
segment["words"][chunk_index:chunk_index + words_count]
|
||||||
):
|
):
|
||||||
timing = original_timing.copy()
|
timing = original_timing.copy()
|
||||||
long_pause = (
|
long_pause = (
|
||||||
|
|||||||
Loading…
x
Reference in New Issue
Block a user