Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / SYSTRAN/faster-whisper issues and pull requests

#1170 - sound-effects missing on output for beam-size > 1

Issue - State: open - Opened by rummyr 6 days ago - 3 comments

#1169 - Version 1.1.0 has onnxruntime thread affinity crash

Issue - State: open - Opened by Appfinity-development 7 days ago - 3 comments

#1168 - Greater error when converted via ctranslate2

Issue - State: open - Opened by hforghani 7 days ago - 2 comments

#1167 - v1.1.0 slower than v1.0.3 on CPU

Issue - State: open - Opened by pablopla 7 days ago

#1165 - how not to get the Smooth text

Issue - State: open - Opened by LRY1994 7 days ago

#1162 - New model: CrisperWhisper

Issue - State: closed - Opened by imsakg 8 days ago - 1 comment

#1161 - Bump version to 1.1.0 and update benchmarks

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago - 2 comments

#1160 - remove `log_prob_low_threshold`

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago

#1160 - remove `log_prob_low_threshold`

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago

#1159 - use `jiwer` instead of `evaluate` in benchmarks

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago

#1159 - use `jiwer` instead of `evaluate` in benchmarks

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago

#1158 - Add new tests

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago

#1158 - Add new tests

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago

#1157 - Fix `list index out of range` in word timestamps

Pull Request - State: closed - Opened by MahmoudAshraf97 9 days ago

#1156 - Long audio needs much more time to transcribe

Issue - State: closed - Opened by terryops 9 days ago - 1 comment

#1155 - Time for a new release?

Issue - State: closed - Opened by RuntimeRacer 9 days ago - 2 comments

#1154 - Add new tests

Pull Request - State: closed - Opened by MahmoudAshraf97 10 days ago

#1154 - Add new tests

Pull Request - State: closed - Opened by MahmoudAshraf97 10 days ago

#1153 - Update vad.py with more sane speech_pad_ms: int = 4000

Pull Request - State: closed - Opened by montvid 10 days ago - 2 comments

#1153 - Update vad.py with more sane speech_pad_ms: int = 4000

Pull Request - State: closed - Opened by montvid 10 days ago - 2 comments

#1152 - how can i not to get the normalized chinese number?

Issue - State: open - Opened by LRY1994 10 days ago - 1 comment

#1150 - Feat: correct type hints

Pull Request - State: open - Opened by Equipo45 12 days ago

#1149 - comparison of Turbo Models

Issue - State: open - Opened by xipingL 12 days ago - 10 comments

#1148 - refactor `multilingual` option

Pull Request - State: closed - Opened by MahmoudAshraf97 13 days ago

#1147 - No need to install cuda driver?

Issue - State: open - Opened by ILG2021 14 days ago - 1 comment

#1146 - Deduplication of language detection functions

Pull Request - State: closed - Opened by MahmoudAshraf97 14 days ago

#1145 - nce

Issue - State: closed - Opened by pierrepv8 14 days ago - 1 comment

#1144 - Whisper Turbo model run on a CPU

Issue - State: closed - Opened by joseph16388 14 days ago - 3 comments

#1143 - Batch size impact on speed and VRAM consumption

Issue - State: closed - Opened by vidalfer 15 days ago - 4 comments

#1142 - add github action to build

Pull Request - State: closed - Opened by BitSteve 15 days ago

#1140 - Some segment has a 1 second shifted after PR #856

Issue - State: closed - Opened by heimoshuiyu 15 days ago

#1139 - Can I serve speechbrain trained model whisper with faster whisper?

Issue - State: open - Opened by cod3r0k 16 days ago - 5 comments

#1138 - Add progress bar

Pull Request - State: closed - Opened by MahmoudAshraf97 16 days ago

#1136 - Added encoder output

Pull Request - State: closed - Opened by SinanAkkoyun 16 days ago - 1 comment

#1135 - Cleanup of `BatchedInferencePipeline`

Pull Request - State: closed - Opened by MahmoudAshraf97 16 days ago - 4 comments

#1134 - change `language_detection_threshold` default value

Pull Request - State: closed - Opened by MahmoudAshraf97 16 days ago

#1132 - Hidden state / embeddings of encoder

Issue - State: closed - Opened by SinanAkkoyun 17 days ago - 2 comments

#1128 - no_speech_prob always returns 0.0.

Issue - State: open - Opened by giaoyyds 18 days ago

#1126 - Use multiple GPUs to process queue

Issue - State: open - Opened by theodufort 19 days ago - 1 comment

#1125 - Update cuda instructions in readme

Pull Request - State: closed - Opened by MahmoudAshraf97 19 days ago

#1123 - cannot import name 'BatchedInferencePipeline'

Issue - State: closed - Opened by acclayer7 22 days ago - 1 comment

#1122 - distill model ignores language

Issue - State: closed - Opened by trigsoft 22 days ago - 1 comment

#1121 - Possible way to integrate support for Intel Iris Xe gpus?

Issue - State: closed - Opened by gimmeAll 22 days ago - 1 comment

#1120 - fasterwhisper and cuda version

Issue - State: closed - Opened by zhuangzhuangliu2345 22 days ago - 7 comments

#1118 - IndexError: list index out of range in add_word_timestamps function

Issue - State: closed - Opened by formater 23 days ago - 10 comments

#1117 - Illegal instruction docker container

Issue - State: open - Opened by danpuz 24 days ago

#1116 - Is this compatible with tinydiarize?

Issue - State: closed - Opened by mathigatti 25 days ago - 1 comment

#1113 - Update feature_extractor.py

Pull Request - State: closed - Opened by BBC-Esq 26 days ago

#1112 - Update audio.py

Pull Request - State: open - Opened by BBC-Esq 26 days ago - 11 comments

#1110 - no

Issue - State: closed - Opened by majisama 28 days ago

#1109 - Custom VAD config with BatchedInferencePipeline

Issue - State: closed - Opened by vidalfer 29 days ago - 1 comment

#1108 - min_silence_duration_ms is not working, Silence detection not working

Issue - State: closed - Opened by andriken 29 days ago - 5 comments

#1107 - CuPy only FE

Pull Request - State: closed - Opened by MahmoudAshraf97 29 days ago

#1106 - Remove `torch` dependency, Faster numpy Feature extraction

Pull Request - State: closed - Opened by MahmoudAshraf97 29 days ago - 21 comments

#1105 - replace `NamedTuple` with `dataclass`

Pull Request - State: closed - Opened by MahmoudAshraf97 about 1 month ago - 6 comments

#1104 - Add Pydantic typing

Pull Request - State: closed - Opened by extrange about 1 month ago - 5 comments

#1102 - Transcription error related to BatchedInferencePipeline and numpy

Issue - State: closed - Opened by shkstar about 1 month ago - 3 comments

#1101 - Use correct features padding for encoder input

Pull Request - State: closed - Opened by MahmoudAshraf97 about 1 month ago - 1 comment

#1100 - Issue with segments when the model doesn't output end time for segment.

Issue - State: open - Opened by bchinnari about 1 month ago - 17 comments

#1099 - Finetuned large-v3 inference problem

Issue - State: open - Opened by sinisha about 1 month ago - 1 comment

#1099 - Finetuned large-v3 inference problem

Issue - State: open - Opened by sinisha about 1 month ago - 14 comments

#1098 - Loop task, transcribe after obtaining new audio files. Eventually, it will boom

Issue - State: closed - Opened by Gitawake about 1 month ago - 2 comments

#1096 - [DRAFT] refactor of transcribe.py[DISCUSS]

Pull Request - State: closed - Opened by BBC-Esq about 1 month ago - 3 comments

#1096 - [DRAFT] refactor of transcribe.py[DISCUSS]

Pull Request - State: open - Opened by BBC-Esq about 1 month ago - 3 comments

#1095 - Audio

Pull Request - State: closed - Opened by BBC-Esq about 1 month ago

#1095 - Audio

Pull Request - State: open - Opened by BBC-Esq about 1 month ago

#1094 - Add modal integration for STT model

Pull Request - State: closed - Opened by amanjain97 about 1 month ago - 1 comment

#1093 - Update tokenizer.py

Pull Request - State: closed - Opened by BBC-Esq about 1 month ago - 2 comments

#1092 - typo: trubo -> turbo

Pull Request - State: closed - Opened by heimoshuiyu about 1 month ago

#1091 - Update audio.py

Pull Request - State: closed - Opened by BBC-Esq about 1 month ago - 1 comment

#1091 - Update audio.py

Pull Request - State: closed - Opened by BBC-Esq about 1 month ago - 1 comment

#1090 - Add support for `turbo` model

Pull Request - State: closed - Opened by MahmoudAshraf97 about 1 month ago

#1086 - CUDA compatibility with CTranslate2

Issue - State: open - Opened by MahmoudAshraf97 about 1 month ago - 4 comments

#1086 - CUDA compatibility with CTranslate2

Issue - State: open - Opened by MahmoudAshraf97 about 1 month ago - 2 comments

#1085 - Running the same code twice giving two different results

Issue - State: open - Opened by bchinnari about 1 month ago - 3 comments

#1084 - Pad audio instead of mel features to reduce word error rates

Pull Request - State: closed - Opened by MahmoudAshraf97 about 1 month ago - 5 comments

#1084 - Pad audio instead of mel features to reduce word error rates

Pull Request - State: open - Opened by MahmoudAshraf97 about 1 month ago - 4 comments

#1083 - Latest CTranslate2 release breaks faster-whisper

Issue - State: closed - Opened by kristopher-smith about 1 month ago - 2 comments

#1083 - Latest CTranslate2 release breaks faster-whisper

Issue - State: closed - Opened by kristopher-smith about 1 month ago - 2 comments

#1081 - KernelRestarter: restarting kernel (1/5), keep random ports

Issue - State: closed - Opened by etb-git1 about 1 month ago - 5 comments

#1081 - KernelRestarter: restarting kernel (1/5), keep random ports

Issue - State: closed - Opened by etb-git1 about 1 month ago - 7 comments

#1080 - cudnn ops64_9.dll is not found

Issue - State: closed - Opened by jhj0517 about 1 month ago - 14 comments

#1079 - Request for help: how to make Whisper results *worse*

Issue - State: closed - Opened by Krzysiu about 1 month ago - 2 comments

#1078 - VOTE HERE - transfer of maintaining faster-whisper repo

Issue - State: closed - Opened by BBC-Esq about 1 month ago - 1 comment

#1077 - Feeding raw audio data to faster whisper over websockets

Issue - State: open - Opened by FredTheNoob about 1 month ago - 1 comment

#1077 - Feeding raw audio data to faster whisper over websockets

Issue - State: open - Opened by FredTheNoob about 1 month ago - 1 comment

#1076 - Hotwords not respected after PR #856

Issue - State: closed - Opened by carolinaxxxxx about 1 month ago - 5 comments

#1076 - Hotwords not respected after PR #856

Issue - State: open - Opened by carolinaxxxxx about 1 month ago - 4 comments