mirror of
https://github.com/openai/whisper.git
synced 2025-11-24 14:35:57 +00:00
update Readme file
This commit is contained in:
parent
daaf0ed4ca
commit
0ae5dcd75b
@ -59,26 +59,3 @@ To confirm accurate transcription:
|
||||
Starting transcription... (Press Ctrl+C to stop)
|
||||
The quick brown fox jumps over the lazy dog
|
||||
```
|
||||
```
|
||||
|
||||
### **Key Improvements**:
|
||||
1. Added **verification steps** to confirm mic is working
|
||||
2. Included **accuracy testing** with standard test sentence
|
||||
3. Added **troubleshooting table** for common issues
|
||||
4. Shows **expected output** example
|
||||
|
||||
### **How to Update**:
|
||||
1. Open `examples/README.md`
|
||||
2. Replace contents with the above markdown
|
||||
3. Commit changes:
|
||||
```bash
|
||||
git add examples/README.md
|
||||
git commit -m "docs: Add detailed verification steps"
|
||||
git push
|
||||
```
|
||||
|
||||
This will help users (including yourself) verify if the transcription is working properly. The test sentence "The quick brown fox..." is particularly useful because:
|
||||
- Contains all English letters
|
||||
- Easy to recognize when correct
|
||||
- Helps identify specific sound recognition issues
|
||||
|
||||
|
||||
@ -36,6 +36,7 @@ def try_standard_rates():
|
||||
continue
|
||||
return None
|
||||
|
||||
|
||||
def try_pulseaudio():
|
||||
"""Force PulseAudio configuration"""
|
||||
try:
|
||||
@ -52,7 +53,7 @@ def try_direct_hw():
|
||||
return None
|
||||
|
||||
# Load model first to fail fast if issues
|
||||
MODEL = whisper.load_model("tiny")
|
||||
MODEL = whisper.load_model("base")
|
||||
|
||||
# Get audio config
|
||||
config = get_audio_config()
|
||||
|
||||
Loading…
x
Reference in New Issue
Block a user