heimoshuiyu/faster-whisper
1
0
Fork 0
Code Issues Pull Requests 1 Packages Projects Releases Wiki Activity
224 Commits 2 Branches 0 Tags
6e42088656e63c4222b8a193e8c783accb2a1df7
Commit Graph

116 Commits

Author SHA1 Message Date
heimoshuiyu
6e42088656 Merge remote-tracking branch 'upstream/master' into prompt 2024-09-04 17:48:06 +08:00
Mahmoud Ashraf
d57c5b40b0 Remove the usage of transformers.pipeline from BatchedInferencePipeline and fix word timestamps for batched inference (#921) 
* fix word timestamps for batched inference

* remove hf pipeline
 2024-07-27 09:02:58 +07:00
zh-plus
83a368e98a Make vad-related parameters configurable for batched inference. (#923) 2024-07-24 09:00:32 +07:00
Jilt Sebastian
eb8390233c New PR for Faster Whisper: Batching Support, Speed Boosts, and Quality Enhancements (#856) 
Batching Support, Speed Boosts, and Quality Enhancements

---------

Co-authored-by: Hargun Mujral <83234565+hargunmujral@users.noreply.github.com>
Co-authored-by: MahmoudAshraf97 <hassouna97.ma@gmail.com>
 2024-07-18 16:48:52 +07:00
heimoshuiyu
4a59bb011d Merge remote-tracking branch 'upstream/master' into prompt 2024-07-10 10:16:35 +08:00
trungkienbkhn
fbcf58bf98 Fix language detection with non-speech audio (#895) 2024-07-05 14:43:45 +07:00
Jordi Mas
1195359984 Filter out non_speech_tokens in suppressed tokens (#898) 
* Filter out non_speech_tokens in suppressed tokens
 2024-07-05 14:43:11 +07:00
ABen
8862bee1f8 Improve language detection when using clip_timestamps (#867) 2024-07-01 16:12:45 +07:00
Napuh
f53be1e811 Add distil models to WhisperModel init and download_model docstrings (#847) 
* chore: add distil models to WhisperModel init docstring and download_model docstring
 2024-05-20 08:51:22 +07:00
Natanael Tan
4acdb5c619 Fix #839 incorrect clip_timestamps being used in model (#842) 
* Fix #839

Changed the code from updating the TranscriptionOptions class instead of the options object which likely was the cause of unexpected behaviour
 2024-05-17 16:35:07 +07:00
Keating Reid
49a80eb8a8 Clarify documentation for hotwords (#817) 
* Clarify documentation for hotwords

* Remove redundant type specifications
 2024-05-06 08:52:59 +07:00
trungkienbkhn
8d5e6d56d9 Support initializing more whisper model args (#807) 2024-05-04 15:12:59 +07:00
jax
847fec4492 Feature/add hotwords (#731) 
* add hotword params

---------

Co-authored-by: jax <jax_builder@gamil.com>
 2024-05-04 15:11:52 +07:00
heimoshuiyu
4ee1d54c14 Merge branch 'master' into prompt 2024-04-08 20:56:49 +08:00
Purfview
b024972a56 Foolproof: Disable VAD if clip_timestamps is in use (#769) 
* Foolproof: Disable VAD if clip_timestamps is in use

Prevent silly things to happen.
 2024-04-02 18:20:34 +02:00
Purfview
8ae82c8372 Bugfix: code breaks if audio is empty (#768) 
* Bugfix: code breaks if audio is empty

Regression since https://github.com/SYSTRAN/faster-whisper/pull/732 PR
 2024-04-02 18:18:12 +02:00
trungkienbkhn
1eb9a8004c Improve language detection (#732) 2024-03-12 15:44:49 +01:00
heimoshuiyu
e50d82c18c Merge remote-tracking branch 'upstream/master' into prompt 2024-03-10 11:53:58 +08:00
Purfview
5090cc9d0d Fix window end heuristic for hallucination_silence_threshold (#706) 
Removes the wishful heuristic causing more issues than it's fixing.

Same as https://github.com/openai/whisper/pull/2043

Example of the issue: https://github.com/openai/whisper/pull/1838#issuecomment-1960041500
 2024-02-29 17:59:32 +01:00
trungkienbkhn
16141e65d9 Add pad_or_trim function to handle segment before encoding (#705) 2024-02-29 17:08:28 +01:00
heimoshuiyu
4b64ef1f70 Merge branch 'master' into prompt 2024-02-23 10:52:53 +08:00
Purfview
30d6043e90 Prevent infinite loop for out-of-bound timestamps in clip_timestamps (#697) 
Same as https://github.com/openai/whisper/pull/2005
 2024-02-22 09:48:35 +01:00
trungkienbkhn
092067208b Add clip_timestamps and hallucination_silence_threshold options (#646) 2024-02-20 17:34:54 +01:00
heimoshuiyu
d04e685ca2 Merge branch 'master' into prompt 2024-02-19 17:31:58 +08:00
Purfview
3aec421849 Add: More clarity of what "max_new_tokens" does (#658) 
* Add: More clarity of what "max_new_tokens" does
 2024-01-28 21:40:33 +01:00
Purfview
00efce1696 Bugfix: Illogical "Avoid computing higher temperatures on no_speech" (#652) 2024-01-24 11:54:43 +01:00
metame
ad3c83045b support distil-whisper (#557) 2024-01-24 10:17:12 +01:00
heimoshuiyu
b835bdaaf1 Merge remote-tracking branch 'upstream/master' into prompt 2023-12-25 17:56:50 +08:00
Purfview
ebcfd6b964 Fix broken prompt_reset_on_temperature (#604) 
* Fix broken prompt_reset_on_temperature

Fixing: https://github.com/SYSTRAN/faster-whisper/issues/603

Broken because `generate_with_fallback()` doesn't return final temperature.

Regression since PR356 -> https://github.com/SYSTRAN/faster-whisper/pull/356
 2023-12-13 13:14:39 +01:00
trungkienbkhn
19329a3611 Word timing tweaks (#616) 2023-12-13 12:38:44 +01:00
Oscaarjs
3084409633 Add V3 Support (#578) 
* Add V3 Support

* update conversion example

---------

Co-authored-by: oscaarjs <oscar.johansson@conversy.se>
 2023-11-24 23:16:12 +01:00
Guillaume Klein
e94711bb5c Add property WhisperModel.supported_languages (#476) 
* Expose function supported_languages

* Make it a method
 2023-09-14 17:42:02 +02:00
Guillaume Klein
a49097e655 Add some missing typing annotations in transcribe.py 2023-09-12 15:45:54 +02:00
Guillaume Klein
81086f6d33 Always run the encoder at the beginning of the loop (#468) 2023-09-12 14:44:37 +02:00
Guillaume Klein
4a41746e55 Log a warning when the model is English-only but the language is set to something else (#454) 2023-09-04 11:55:40 +02:00
Guillaume Klein
1e6eb967c9 Add "large" alias for "large-v2" model (#453) 2023-09-04 11:54:42 +02:00
Guillaume Klein
f0ff12965a Expose generation parameter no_repeat_ngram_size (#449) 2023-09-01 17:31:30 +02:00
MinorJinx
e87fbf8a49 Added audio duration after VAD to TranscriptionInfo object (#445) 
* Added VAD removed audio duration to TranscriptionInfo object

Along with the duration of the original audio, this commit  adds the seconds of audio removed by the VAD to the returned info obj

* Chaning naming for duration_after_vad

Instead of the property returning the audio duration removed, it now returns the final duration after the vad.
If vad_filter is False or if it doesn't remove any audio, the original duration is returned.
 2023-08-31 17:19:48 +02:00
Aisu Wata
1562b02345 added repetition_penalty to TranscriptionOptions (#403) 
Co-authored-by: Aisu Wata <aisu.wata0@gmail.com>
 2023-08-06 10:08:24 +02:00
Purfview
1ce16652ee Adds DEBUG log message for prompt_reset_on_temperature (#399) 
Produce DEBUG log message if prompt_reset_on_temperature threshold is met.
 2023-08-04 09:06:17 +02:00
Purfview
857be6f621 Rename clear_previous_text_on_temperature argument (#398) 
`prompt_reset_on_temperature` is more clear what it does.
 2023-08-03 18:44:37 +02:00
KH
1a1eb1a027 Add clear_previous_text_on_temperature parameter (#397) 
* Add clear_previous_text_on_temperature parameter

* Add a description
 2023-08-03 15:40:58 +02:00
Guillaume Klein
0f55c436fe Invalidate the cached encoder output when no_speech threshold is met (#376) 2023-07-24 10:57:15 +02:00
KH
e786e26f75 Return result with best log prob when all temperature fallbacks failed (#356) 
* Resolve Inference Selection Bug

* Refactor for better readability

* Filter out results with compression_ratio

* Refactor to avoid variable repetition

* Fix incorrect index and perform minor refactoring

* Remove final_temperature variable
 2023-07-20 16:13:11 +02:00
KH
687db319e0 Remove duplicate code (#359) 2023-07-18 16:03:01 +02:00
Guillaume Klein
0e051a5b77 Prepend prefix tokens with the initial timestamp token (#358) 2023-07-18 15:22:39 +02:00
Hoon
3b4a6aa1c2 Improve timestamp heuristics (#336) 
* Improve timestamp heuristics

* Chore
 2023-07-05 15:16:53 +02:00
zh-plus
c7cb2aa8d4 Add support for using whisper models from Huggingface by specifying the model id. (#334) 
* Add support for downloading CTranslate-converted models from Huggingface.

* Update utils.py to pass Flake8.

* Update utils.py to pass black.

* Remove redundant usage instructions.

* Apply suggestions from code review

Co-authored-by: Guillaume Klein <guillaumekln@users.noreply.github.com>

---------

Co-authored-by: Guillaume Klein <guillaumekln@users.noreply.github.com>
 2023-07-03 17:40:10 +02:00
Guillaume Klein
c0d93d0829 Avoid computing higher temperatures on no_speech segments (#225) 
Port commit e334ff141d
 2023-07-03 10:20:36 +02:00
Guillaume Klein
19c294f978 Squash long words at window and sentence boundaries (#226) 
Port commit 255887f219
 2023-07-03 10:20:20 +02:00