Commit Graph

9 Commits

Author SHA1 Message Date
Patrick von Platen
7a787c68c6
[Speech models] Disable non-existing chunking in tests (#15163) 2022-01-16 17:15:19 +01:00
Patrick von Platen
600496fa50
[Wav2Vec2] Rename model's feature extractor to feature encoder (#14959)
* rename classes

* clean up more namings

* remove bogus file

* Apply suggestions from code review

* Apply suggestions from code review

* replace more names

* more regex replace

* make style

* correct

* correct more

* make style

* finish

* correct more in wav2vec2

* make style

* improve freeze_extractor

* add aliases

* add tf aliases
2021-12-28 20:33:23 +01:00
Anton Lozhkov
ee55ea692b
Update diarization and WavLM tolerances (#14902) 2021-12-23 19:53:56 +03:00
Anton Lozhkov
3883e3a75e
Add SD and SV heads for WavLM (#14847)
* Add converted heads

* Add dummies
2021-12-20 16:40:56 +03:00
Lysandre Debut
d194d639ab
Remove datasets requirement (#14795) 2021-12-16 14:34:14 -05:00
Anton Lozhkov
48463ebb33
Add Speaker Diarization and Verification heads (#14723)
* Models

* Squashed commit of the following:

commit 72278e1e931a16d0879acc77f65762f3364833d0
Author: anton-l <aglozhkov@gmail.com>
Date:   Fri Dec 10 21:45:08 2021 +0300

* Add unispeech heads

* Add sd/sv automodels

* Docs cleanup

* Fix docstrings

* rename xvector classes

* examples

* Tests cleanup

* Style

* Better checkpoints for tests

* leftover docs

* apply review suggestions

* Style + init tests

* Update unispeech-sat tdnn downsampling
2021-12-16 19:22:14 +03:00
Sylvain Gugger
040fd47162
Fix gradient_checkpointing backward compatibility (#14408)
* Fix gradient_checkpointing backward compatibility

* Remove needless line

* make sure mask prob is big enough and length small enough

* Fix tests

Co-authored-by: patrickvonplaten <patrick.v.platen@gmail.com>
2021-11-16 08:58:42 -05:00
Yih-Dar
9450bfcc6c
Add more missing models to models/__init__.py (#14177)
* Add missing models to models/__init__.py

* Fix issues previously undetected

* Add UniSpeechSatForPreTraining to all_model_classes

* fix unispeech sat

* fix

* Add check_model_list() to check_repo.py

* Remove _ignore_models = ["bort"]

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
Co-authored-by: patrickvonplaten <patrick.v.platen@gmail.com>
2021-11-01 10:52:36 +00:00
Patrick von Platen
9f3aa46f45
Add Unispeech & Unispeech-SAT (#13963)
* unispeech

* add copy from

* remove hubert copy from

* finish for today

* add unispeech-sat

* adapt more

* up

* up

* up

* up

* add modeling

* add tests

* up

* up

* finish

* up

* Apply suggestions from code review

* up

* up

* Apply suggestions from code review

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>

* up

* up

Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
2021-10-26 18:59:58 +02:00