MediaChef

Any media into any.
Speech into text.All on your computer.

MediaChef is a free desktop app that turns FFmpeg and Whisper into simple recipe cards: convert video and audio, extract sound, make subtitles — with no file uploads, no size limits, no queues.

Free and open source — view the code on GitHub

Open source · GPL-3.0Works offlineEnglish & RussianmacOS · Windows · Linux
Drop files here

How it works

1

Drop a file

Drag in a video or a recording — MediaChef reads what it is and shows only the recipes that fit.

2

Pick a recipe

Every action is a plain card: “Video → MP3”, “Subtitles for video”, “Compress to 25 MB”. Sensible defaults, no flags to learn.

3

Take the result

The file lands next to the original — or in a folder you choose. A clear queue shows progress and time left.

Recipes instead of commands

FFmpeg can do almost anything — in the language of the terminal. MediaChef translates: you choose what to do, the parameters are already set. A live preview even shows the real command, so you learn as you go.

Extract audio to MP3

Pull the soundtrack out of any video.

Subtitles for a video

Whisper listens and writes an SRT file.

Compress a video

Fit a clip into a messenger-friendly size.

Video → GIF

A sharp looping GIF with tuned palette.

Transcribe a recording

Meeting or voice memo into plain text.

Translate speech to English

Whisper transcribes and translates in one pass.

MP4 → MKV, lossless

Repackage without re-encoding — instant.

Mute a video

Drop every audio track, keep the picture.

Speech-to-text that stays on your device

Whisper, running locally. OpenAI's speech model executes on your processor — recordings never leave the machine.

Models downloaded in-app. From the fast tiny to the accurate large-v3-turbo — pick per task on the Models screen.

Text or subtitles. Plain TXT, SRT and VTT with timestamps, or JSON for tooling.

Honest results. If a file has no speech, MediaChef says “No speech detected” — it never ships an empty green checkmark.

00:00:04 Hi everyone, let's get started… 00:00:11 First topic — plans for the quarter. 00:00:19 There are three scenarios, sharing now. .srt · .txt

Your files never travel

Online converters ask you to upload a file to someone else's server, wait in a queue and trust their retention policy. MediaChef works on your CPU: a gigabyte video or a private meeting recording converts the same way — even with Wi-Fi off.

No uploadsNo size limitsNo subscriptions

Open source, batteries included

GPL-3.0, with the full development history public on GitHub. The app ships FFmpeg and Whisper inside — download, open, use.

$ ffmpeg -i input.mp4 -vn -b:a 192k output.mp3
# MediaChef types this for you

Questions, answered

Is it really free?

Yes. MediaChef is open source under GPL-3.0. The full converter and transcription are free; the code is public on GitHub.

Which formats are supported?

Everything FFmpeg reads — MP4, MKV, MOV, WebM, AVI, MP3, WAV, FLAC, M4A, OGG and many more. Recipes cover the everyday conversions; an expert card accepts any FFmpeg arguments.

Where do my files get uploaded?

Nowhere. Conversion and transcription run entirely on your computer. The only thing MediaChef ever downloads is a Whisper model — once, from the Models screen.

How accurate is the transcription?

It uses OpenAI's Whisper models. Accuracy scales with the model you pick: tiny is instant and rough, large-v3-turbo is near-human on clear speech. Language is detected automatically.

Does it work offline?

Yes. After you download a speech model once, everything — converting and transcribing — works with no internet at all.

Why not just use an online converter?

For a small public file — sure. But private recordings, large videos, batches and anything confidential are better done locally: faster than uploading, no limits, nothing stored on someone's server.

Put a chef in charge of your media

Free, open source, three platforms.

MediaChef is young: builds are not yet signed by Apple/Microsoft, so the first launch asks for confirmation — a plain-text how-to ships inside every download.