Commit Graph

  • ed9a06cd89 Adds new VAD parameters (#1386) master Purfview 2025-11-19 14:40:46 +00:00
  • 2eeafe05de Update Silero-VAD weights to v6.2 (#1390) Purfview 2025-11-19 14:14:42 +00:00
  • cf42429f96 Remove "local_dir_use_symlinks" from download_model() (#1389) Purfview 2025-11-18 18:59:01 +00:00
  • 65882eee9f Bump version to 1.2.1 v1.2.1 Mahmoud Ashraf 2025-10-31 14:31:14 +03:00
  • 409a6919f9 Prevent timestamps restoration when clip timestamps are provided in batched inference (#1376) Mahmoud Ashraf 2025-10-31 14:26:17 +03:00
  • 00a5b26b1f Offload retry logic to hf hub (#1382) Mahmoud Ashraf 2025-10-30 22:11:01 +03:00
  • ba812f55a2 Fix quotes for Python version in CI workflow 310 Mahmoud Ashraf 2025-10-30 21:14:30 +03:00
  • 44466c7535 Upgrade Python version from 3.9 to 3.10 in CI Mahmoud Ashraf 2025-10-30 21:12:36 +03:00
  • e3e46675b2 Update Python version requirements to 3.10 and 3.12 Mahmoud Ashraf 2025-10-30 21:11:50 +03:00
  • 14ad587c98 Update Python version requirement to 3.10 or greater Mahmoud Ashraf 2025-10-30 21:11:07 +03:00
  • 9090997d25 Fix a typo (#1377) Purfview 2025-10-22 13:51:56 +01:00
  • dea24cbcc6 Upgrade to Silero-VAD V6 (#1373) Mahmoud Ashraf 2025-10-14 15:29:56 +03:00
  • 14ba1051f3 Fix: add <|nocaptions|> to suppressed tokens (#1338) Mario 2025-10-10 20:56:54 +02:00
  • c26d609974 only merge when clip_timestamps are not provided (#1345) Mahmoud Ashraf 2025-08-16 14:30:50 +03:00
  • 4bd98d5c5b Update README.md to include whisper-fastapi (#1325) 黑墨水鱼 2025-08-11 18:44:48 +08:00
  • 93001a9438 bump version to 1.2.0 v1.2.0 Mahmoud Ashraf 2025-08-06 03:31:36 +03:00
  • a0c3cb9802 Remove Silence in Batched transcription (#1297) Mahmoud Ashraf 2025-08-06 03:30:59 +03:00
  • fbeb1ba731 get correct index for samples (#1336) Mahmoud Ashraf 2025-08-06 03:17:45 +03:00
  • d3bfd0a305 feat: Allow loading of private HF models (#1309) Rishil 2025-06-02 12:12:34 +01:00
  • 43d4163fe0 Support distil-large-v3.5 (#1311) Mahmoud Ashraf 2025-06-02 14:09:20 +03:00
  • 700584b2e6 feat: allow passing specific revision to download (#1292) Felix Mosheev 2025-04-30 00:55:48 +03:00
  • 1383fd4d37 Update README.md with speaches instead of faster-whisper-server (#1267) David Jiménez 2025-03-20 15:20:26 +01:00
  • 9e657b47cb Bump version to 1.1.1 v1.1.1 Mahmoud Ashraf 2025-01-01 17:44:54 +03:00
  • 11fd8ab301 Fix neg_threshold (#1191) Purfview 2024-12-29 11:38:58 +00:00
  • 95164297ff Add duration of audio and VAD removed duration to BatchedInferencePipeline (#1186) Dragoș Bălan 2024-12-23 16:23:40 +01:00
  • 1b24f284c9 Reduce VAD memory usage (#1198) Purfview 2024-12-12 12:23:30 +00:00
  • b568faec40 Add Open-dubbing into community projects (#1034) Jordi Mas 2024-12-12 11:36:04 +01:00
  • f32c0e8af3 Make batched suppress_tokens behaviour same as in sequential (#1194) Purfview 2024-12-11 11:51:38 +00:00
  • 8327d8cc64 Brings back original VAD parameters naming (#1181) Purfview 2024-12-01 17:41:53 +00:00
  • 22a5238b56 Upgrade CI to 3.9 and drop Python 3.8 support(#1184) Mahmoud Ashraf 2024-12-01 19:38:27 +02:00
  • 97a4785fa1 Bump version to 1.1.0 and update benchmarks (#1161) v1.1.0 Mahmoud Ashraf 2024-11-21 18:22:01 +02:00
  • 08f6900217 remove log_prob_low_threshold (#1160) Mahmoud Ashraf 2024-11-20 23:03:21 +02:00
  • 9c8ef76c98 use jiwer instead of evaluate in benchmarks (#1159) Mahmoud Ashraf 2024-11-20 22:51:55 +02:00
  • 491852e1b9 Add new tests (#1158) Mahmoud Ashraf 2024-11-20 13:50:57 +02:00
  • f830c6f241 Fix list index out of range in word timestamps (#1157) Mahmoud Ashraf 2024-11-20 12:36:58 +02:00
  • bcd8ce0fc7 refactor multilingual option (#1148) Mahmoud Ashraf 2024-11-19 23:14:59 +02:00
  • be9fb36ed3 Cleanup of BatchedInferencePipeline (#1135) Mahmoud Ashraf 2024-11-17 15:45:32 +02:00
  • a6f8fbae00 Refactor of language detection functions (#1146) Mahmoud Ashraf 2024-11-16 12:53:07 +02:00
  • 53bbe54016 fix: Use correct seek value in output, fix word timestamps when the initial timestamp is not zero (#1141) 黑墨水鱼 2024-11-15 19:57:38 +08:00
  • 85e61ea111 Add progress bar to WhisperModel.transcribe (#1138) Mahmoud Ashraf 2024-11-14 16:12:39 +02:00
  • 3e0ba86571 Remove torch dependency, Faster numpy Feature extraction (#1106) Mahmoud Ashraf 2024-11-14 11:57:10 +02:00
  • 8f01aee36b Update WhisperModel documentation to list all available models (#1137) Mahmoud Ashraf 2024-11-13 18:26:01 +02:00
  • c2bf036234 change language_detection_threshold default value (#1134) Mahmoud Ashraf 2024-11-13 16:07:46 +02:00
  • fb65cd387f Update cuda instructions in readme (#1125) Mahmoud Ashraf 2024-11-12 14:51:26 +02:00
  • 203dddb047 replace NamedTuple with dataclass (#1105) Mahmoud Ashraf 2024-11-05 11:32:20 +02:00
  • 814472fdbf Revert CPU default threads to 0 Mahmoud Ashraf 2024-10-30 23:00:36 +03:00
  • f978fa2979 Revert CPU default threads to 4 (#965) Ozan Caglayan 2024-10-30 13:50:49 +00:00
  • 2386843fd7 Use correct features padding for encoder input (#1101) Mahmoud Ashraf 2024-10-29 17:58:05 +03:00
  • c2a1da1bd9 typo: trubo -> turbo (#1092) 黑墨水鱼 2024-10-26 05:28:16 +08:00
  • b2da05582c Add support for turbo model (#1090) Mahmoud Ashraf 2024-10-25 15:50:23 +03:00
  • 2dbca5e559 Use Silero VAD in Batched Mode (#936) Mahmoud Ashraf 2024-10-24 12:05:25 +03:00
  • 574e2563e7 Update Dockerfile to ensure compatibility with CT2==4.5.0 Mahmoud Ashraf 2024-10-23 18:28:27 +03:00
  • 42b8681edb revert back to using PyAV instead of torchaudio (#961) Mahmoud Ashraf 2024-10-23 15:26:18 +03:00
  • d57c5b40b0 Remove the usage of transformers.pipeline from BatchedInferencePipeline and fix word timestamps for batched inference (#921) Mahmoud Ashraf 2024-07-27 05:02:58 +03:00
  • 83a368e98a Make vad-related parameters configurable for batched inference. (#923) zh-plus 2024-07-24 10:00:32 +08:00
  • eb8390233c New PR for Faster Whisper: Batching Support, Speed Boosts, and Quality Enhancements (#856) Jilt Sebastian 2024-07-18 11:48:52 +02:00
  • fbcf58bf98 Fix language detection with non-speech audio (#895) trungkienbkhn 2024-07-05 14:43:45 +07:00
  • 1195359984 Filter out non_speech_tokens in suppressed tokens (#898) Jordi Mas 2024-07-05 09:43:11 +02:00
  • c22db5125d Bump version to 1.0.3 (#887) v1.0.3 trungkienbkhn 2024-07-01 16:36:12 +07:00
  • 8862bee1f8 Improve language detection when using clip_timestamps (#867) ABen 2024-07-01 17:12:45 +08:00
  • 8d400e9870 Upgrade to Silero-Vad V5 (#884) Ki Hoon Kim 2024-07-01 17:40:37 +09:00
  • bced5f04c0 docs: add 'faster-whisper-server' community integration (#861) Fedir Zadniprovskyi 2024-06-05 08:27:41 -07:00
  • 65551c081f Docker file improvements (#848) Fedir Zadniprovskyi 2024-05-19 19:13:19 -07:00
  • f53be1e811 Add distil models to WhisperModel init and download_model docstrings (#847) Napuh 2024-05-20 03:51:22 +02:00
  • 4acdb5c619 Fix #839 incorrect clip_timestamps being used in model (#842) Natanael Tan 2024-05-17 17:35:07 +08:00
  • a1c3583c96 Update README.md (#841) Peter Krantz 2024-05-17 10:24:47 +02:00
  • 2036d12634 Add Dockerfile example (#828) trungkienbkhn 2024-05-13 16:33:09 +07:00
  • 2f6913efc8 Bump version to 1.0.2 (#816) v1.0.2 trungkienbkhn 2024-05-06 09:02:54 +07:00
  • e11d58599d Allow av to include version 12. (#819) ddorian 2024-05-06 03:57:35 +02:00
  • 49a80eb8a8 Clarify documentation for hotwords (#817) Keating Reid 2024-05-05 21:52:59 -04:00
  • 8d5e6d56d9 Support initializing more whisper model args (#807) trungkienbkhn 2024-05-04 15:12:59 +07:00
  • 6eec07739e Add benchmarking logic for memory, wer and speed (#773) trungkienbkhn 2024-05-04 15:12:43 +07:00
  • 847fec4492 Feature/add hotwords (#731) jax 2024-05-04 16:11:52 +08:00
  • 46080e584e Loosening tokenizers version constraint (#804) Keating Reid 2024-05-04 04:10:24 -04:00
  • 3d1de60ef3 CUDA version and updated installation instructions (#785) Sidharth Rajaram 2024-05-04 01:09:59 -07:00
  • 91c8307aa6 make faster_whisper.assets as a valid python package to distribute (#772) (#774) otakutyrant 2024-04-03 00:22:22 +08:00
  • b024972a56 Foolproof: Disable VAD if clip_timestamps is in use (#769) Purfview 2024-04-02 17:20:34 +01:00
  • 8ae82c8372 Bugfix: code breaks if audio is empty (#768) Purfview 2024-04-02 17:18:12 +01:00
  • e0c3a9ed34 Update project github link to SYSTRAN (#746) trungkienbkhn 2024-03-27 14:31:17 +07:00
  • a67e0e47ae Add support for distil-large-v3 (#755) Sanchit Gandhi 2024-03-26 13:58:39 +00:00
  • 1eb9a8004c Improve language detection (#732) trungkienbkhn 2024-03-12 21:44:49 +07:00
  • a342b028b7 Bump version to 1.0.1 (#725) v1.0.1 trungkienbkhn 2024-03-01 17:32:12 +07:00
  • 5090cc9d0d Fix window end heuristic for hallucination_silence_threshold (#706) Purfview 2024-02-29 16:59:32 +00:00
  • 09cd57e7f3 Fix typo 'ditil' (#721) Gabriel F 2024-02-29 13:08:58 -03:00
  • 16141e65d9 Add pad_or_trim function to handle segment before encoding (#705) trungkienbkhn 2024-02-29 23:08:28 +07:00
  • 2b1d8cc69b bump version 0.10.1 to fix broken 0.10.0 v0.10.1 v0.10.1 Dang Chuan Nguyen 2024-02-22 13:03:43 +01:00
  • 06d32bf0c1 Bump version to 1.0.0 (#696) v1.0.0 trungkienbkhn 2024-02-22 15:49:01 +07:00
  • 30d6043e90 Prevent infinite loop for out-of-bound timestamps in clip_timestamps (#697) Purfview 2024-02-22 08:48:35 +00:00
  • 22c75d0cc3 Update README.md (#672) BBC-Esq 2024-02-21 04:18:11 -05:00
  • 092067208b Add clip_timestamps and hallucination_silence_threshold options (#646) trungkienbkhn 2024-02-20 23:34:54 +07:00
  • 6ffcbdfbc2 Fix typos in README.md (#668) Jordi Mas 2024-02-20 17:33:17 +01:00
  • 52695567c9 Bumps up PyAV version to support Python 3.12.x (#679) Purfview 2024-02-20 16:31:07 +00:00
  • c6b28ed3a0 Update README.md (#685) IlianP 2024-02-20 17:28:00 +01:00
  • 4ab646035f Upgrade ctranslate2 version to support CUDA 12 (#694) trungkienbkhn 2024-02-20 23:26:55 +07:00
  • f144e4c83d Expands the note for distil-whisper (#659) Purfview 2024-01-28 20:48:40 +00:00
  • 3aec421849 Add: More clarity of what "max_new_tokens" does (#658) Purfview 2024-01-28 20:40:33 +00:00
  • 64b9f244bd Whisper-Streaming mention (#656) Dominik Macháček 2024-01-25 18:27:27 +01:00
  • 00efce1696 Bugfix: Illogical "Avoid computing higher temperatures on no_speech" (#652) Purfview 2024-01-24 10:54:43 +00:00
  • ad3c83045b support distil-whisper (#557) metame 2024-01-24 17:17:12 +08:00
  • 72ff979a2e Add GUI faster-whisper project README.md (#554) Jürgen Fleiß 2024-01-18 13:01:02 +01:00