whishper

Why we included this project

Whishper takes Whisper and makes it usable outside the command line. Upload an audio or video file, or paste a URL that yt-dlp can fetch, and you get transcripts you can export as TXT, VTT, SRT, or JSON. The subtitle editor handles segment splitting and insertion, warns on high characters-per-second rates, and can translate captions via LibreTranslate without leaving the browser. Everything runs locally on the FasterWhisper backend, so it works offline, stays responsive on CPU, and can lean on an NVIDIA GPU when you need speed. That makes it a solid pick for teams transcribing interviews, podcasts, lectures, or video captions who would rather not send that audio to a third party. Note that the author is mid-way through a v4 rewrite, so the current branch is not receiving new releases.