Jong Wook Kim 4179ed2475 add large-v2 model ...

- The "large-v2" model is trained for more epochs with regularization and shows improved performance compared to the previous large.
- It has the same architecture as the original large model.
- When `load_model("large")` is called, the "large-v2" model will be loaded.
- We will soon update the paper regarding this new model.

2022-12-05 11:07:14 -05:00

..

assets

initial commit

2022-09-22 01:09:43 +09:00

normalizers

initial commit

2022-09-22 01:09:43 +09:00

__init__.py

add large-v2 model

2022-12-05 11:07:14 -05:00

__main__.py

initial commit

2022-09-22 01:09:43 +09:00

audio.py

Fix bug (#305)

2022-10-17 11:38:20 -07:00

decoding.py

suppress generating non-timestamp tokens at the beginning (#532)

2022-11-15 11:44:36 -08:00

model.py

invoking __call__ instead of forward()

2022-11-16 04:18:50 -08:00

tokenizer.py

allow hyphens and single quotes between words

2022-09-23 20:11:27 +09:00

transcribe.py

fix to return only the text token ids

2022-11-15 16:25:11 -08:00

utils.py

fix compression ratio function (#561)

2022-12-04 17:27:42 -06:00