Purfview

Systematic cases would be: The false-positives on ~silence. (introduced in v4) Inaccurate end of segments, trailing usually includes up to ~1000 ms of "padding". (introduced in v4) Maybe not systematic...

Those distil models are for English language only.

> increase dialogue volume Try `--ff_speechnorm` in faster-whisper.

> Hi, @Purfview, > > Could you possibly shed light on some queries [here](https://github.com/URUWorks/TeroSubtitler/discussions/67), please? On what specific queries? Btw, thx for donation.

>Regarding the list of systems in the top post... That's about cpp,,, >Also, I wondered if you have thoughts on https://github.com/URUWorks/TeroSubtitler/issues/256. Try `--standard`

Offtopic question: Is it possible to get VAD results faster, for example [pyannote-onnx ](https://github.com/pengzhendong/pyannote-onnx) implementation is ~5 times faster for me? >Would you mind opening a PR removing the mention...

That site is added to https://github.com/Purfview/IMDb-Scout-Mod

Distil models are not multilingual, it's English only.

> Hi, I have been working on faster whisper and trying to use the distil-whisper model. However, distil-whisper supports 30s of audio chunks and using it with faster whisper only...