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

#77 - Fix variable name reference

Pull Request - State: closed - Opened by jordimas over 1 year ago - 1 comment

#76 - Transcribe results are not deterministic in diferent executions

Issue - State: closed - Opened by jordimas over 1 year ago - 2 comments

#75 - CUDA failed with error invalid argument

Issue - State: closed - Opened by dodysw2 over 1 year ago - 6 comments

#74 - av audio library can not determine audio format

Issue - State: closed - Opened by xanguera over 1 year ago - 8 comments

#73 - Run the encoder only once for each 30-second window

Pull Request - State: closed - Opened by guillaumekln over 1 year ago

#72 - Language specification of 'EN' not working

Issue - State: closed - Opened by madalinstan-greppysystems over 1 year ago - 5 comments

#71 - Windows process crashes when the GPU model is unloaded

Issue - State: open - Opened by satisl over 1 year ago - 48 comments

#70 - Automatically download converted models from the Hugging Face Hub

Pull Request - State: closed - Opened by guillaumekln over 1 year ago

#69 - Cannot produce the same result

Issue - State: closed - Opened by zara0m over 1 year ago - 9 comments

#68 - Run some automatic tests with GitHub Actions

Pull Request - State: closed - Opened by guillaumekln over 1 year ago

#67 - How to improve speed of iterataing over segments?

Issue - State: closed - Opened by zara0m over 1 year ago - 8 comments

#66 - This is very cool, but push to even higher gpu usage?

Issue - State: closed - Opened by junchen6072 over 1 year ago - 9 comments

#65 - float16 does not appear to work on CPU with fp16 capabilities

Issue - State: closed - Opened by FlippFuzz over 1 year ago - 14 comments

#64 - Can't use it on gpu

Issue - State: closed - Opened by satisl over 1 year ago - 8 comments

#63 - how can I use this repo? The instructions are not so clear for me

Issue - State: closed - Opened by search620 over 1 year ago - 3 comments

#62 - Punctuation missed comparing to original Whisper

Issue - State: closed - Opened by marlon-br over 1 year ago - 3 comments

#61 - utf8 decoding bug

Issue - State: closed - Opened by davidp-10 over 1 year ago - 3 comments

#60 - Is possible to publish a package at https://pypi.org/

Issue - State: closed - Opened by jordimas over 1 year ago - 2 comments

#59 - batch execution transcribe in faster-whisper

Issue - State: closed - Opened by eschmidbauer over 1 year ago - 38 comments

#58 - Segmentation fault on word_timestamps using custom model

Issue - State: closed - Opened by giannhskp over 1 year ago - 3 comments

#57 - out of bounds using word timestamps

Issue - State: closed - Opened by igiotis over 1 year ago - 2 comments

#56 - 'Tokenizer' object has no attribute 'language_code'

Issue - State: closed - Opened by laphang over 1 year ago - 2 comments

#55 - Colab

Issue - State: closed - Opened by ryoppippi over 1 year ago - 2 comments

#54 - Support for variable size chunks

Issue - State: closed - Opened by vackosar over 1 year ago - 3 comments

#53 - Memory spike at the end of transcription

Issue - State: closed - Opened by vackosar over 1 year ago - 1 comment

#52 - Word confidence scores

Issue - State: closed - Opened by David-Pye over 1 year ago - 4 comments

#51 - Apple Neural Engine on M1 Macs?

Issue - State: closed - Opened by maxcribe over 1 year ago - 3 comments

#50 - IndexError: index 1 is out of bounds for axis 0 with size 1

Issue - State: closed - Opened by haveyouwantto over 1 year ago - 9 comments

#47 - Incompatible with CUDA v12.1

Issue - State: closed - Opened by pandruszkow over 1 year ago - 8 comments

#46 - ctranslate2 version 3.9.0 has an error

Issue - State: closed - Opened by sergioudi over 1 year ago - 3 comments

#45 - inference slowed down with `word_timestamps=True`

Issue - State: closed - Opened by eschmidbauer over 1 year ago - 3 comments

#44 - Inference on long files

Issue - State: closed - Opened by databill86 over 1 year ago - 8 comments

#43 - Add word-level timestamps

Pull Request - State: closed - Opened by guillaumekln over 1 year ago

#41 - Zaz frances wrong

Issue - State: closed - Opened by anthonyanikos over 1 year ago - 10 comments

#40 - Repetited text within and between segments

Issue - State: closed - Opened by cxumol over 1 year ago - 7 comments

#39 - Feature : Add support for VAD filter

Issue - State: closed - Opened by acul3 over 1 year ago - 15 comments

#38 - Slower than original Whisper on ARM 64bit (Raspberry Pi 4/Orange Pi 5)

Issue - State: closed - Opened by fquirin over 1 year ago - 2 comments

#37 - RuntimeError: No SGEMM backend on CPU

Issue - State: closed - Opened by Jeronymous over 1 year ago - 3 comments

#36 - If anybody shared the translated models?

Issue - State: closed - Opened by bigbenbird over 1 year ago - 5 comments

#35 - transcription speed blowouts

Issue - State: closed - Opened by ooobo over 1 year ago - 2 comments

#34 - Exception: Model "openai/whisper-tiny.en" on the Hub doesn't have a tokenizer

Issue - State: closed - Opened by mattaltberg over 1 year ago - 2 comments

#33 - How to assign more CPU on this python script

Issue - State: closed - Opened by B1gM8c over 1 year ago - 7 comments

#32 - How do I run this?

Issue - State: closed - Opened by ghost over 1 year ago - 4 comments

#31 - how to update faster-whisper?

Issue - State: closed - Opened by harry0703 over 1 year ago - 1 comment

#30 - What files need to be prepared to convert my own model

Issue - State: closed - Opened by kFoodie over 1 year ago - 1 comment

#29 - word_timestamps on Faster Whisper

Issue - State: closed - Opened by playdasegunda over 1 year ago - 1 comment

#28 - Crosstalk when multithreading.

Issue - State: closed - Opened by sphinxrave over 1 year ago - 3 comments

#26 - Why is gpu slower than cpu

Issue - State: closed - Opened by kFoodie over 1 year ago - 15 comments

#25 - Thank you for creating this ,This is very fast

Issue - State: closed - Opened by skillhacker-code over 1 year ago - 2 comments

#24 - Two different errors with converting MP3s to WAV

Issue - State: closed - Opened by emhagman over 1 year ago - 6 comments

#23 - Is medium model use the same tiny tokernizer file?

Issue - State: closed - Opened by hawgjmrd72 over 1 year ago - 1 comment

#22 - decode_audio error

Issue - State: closed - Opened by Latrolage over 1 year ago - 3 comments

#21 - Accept the audio waveform as an input to transcribe()

Pull Request - State: closed - Opened by guillaumekln almost 2 years ago

#20 - Add support for passing buffers to transcribe

Issue - State: closed - Opened by royce-mathew almost 2 years ago - 3 comments

#19 - the timestamps of the texts are not accuracy.

Issue - State: closed - Opened by jackleibest almost 2 years ago - 5 comments

#18 - Segmentation fault on Mac M1 during conversion

Issue - State: closed - Opened by deusdeorum0 almost 2 years ago - 3 comments

#17 - Any Colab notebook to test?

Issue - State: closed - Opened by stevevaius2015 almost 2 years ago - 2 comments

#16 - Any ideas to determine if the pronunciation is clear or fluent ?

Issue - State: closed - Opened by sirolei almost 2 years ago - 1 comment

#15 - Shorter segments?

Issue - State: closed - Opened by ronyfadel almost 2 years ago - 7 comments

#14 - Support float32?

Issue - State: closed - Opened by hippylover almost 2 years ago - 2 comments

#13 - TypeError when providing language to model

Issue - State: closed - Opened by abramovi almost 2 years ago - 2 comments

#12 - word-level timestamps

Issue - State: closed - Opened by eschmidbauer almost 2 years ago - 9 comments

#11 - Different output from mono audio and stereo audio

Issue - State: closed - Opened by zanjabil2502 almost 2 years ago - 3 comments

#10 - Different output for `medium` between openai/whisper and this

Issue - State: closed - Opened by emhagman almost 2 years ago - 33 comments

#9 - Speed tests and comparison to other Whisper versions

Issue - State: closed - Opened by fquirin almost 2 years ago - 25 comments

#8 - Mac M1, compute type is float32 . how can we transform a float32 model?

Issue - State: closed - Opened by sirolei almost 2 years ago - 3 comments

#7 - How to wait for function to finish

Issue - State: closed - Opened by ItakeLs almost 2 years ago - 2 comments

#6 - beam_size default setting

Issue - State: closed - Opened by kristopher-smith almost 2 years ago - 1 comment

#5 - Expose the device_index argument

Pull Request - State: closed - Opened by guillaumekln almost 2 years ago

#4 - How to select a GPU?

Issue - State: closed - Opened by EtienneAb3d almost 2 years ago - 1 comment

#3 - A part of the beginning of my audio was cut

Issue - State: closed - Opened by ubanning almost 2 years ago - 7 comments

#2 - Add the initial_prompt parameter

Pull Request - State: closed - Opened by guillaumekln almost 2 years ago

#1 - initial_prompt?

Issue - State: closed - Opened by EtienneAb3d almost 2 years ago