Does it handle Québécois French?

Yes, along with Metropolitan, Belgian and Swiss French. Strong regional slang is where accuracy dips, as it does in any language.

Are accents and punctuation included?

Yes — é, è, ê, ç and proper punctuation are produced automatically, so the transcript reads as written French.

Can I export subtitles?

Yes, as .SRT or .VTT with timestamps, ready for a video editor or YouTube.

Do numbers and dates come out as digits or words?

Mostly as digits, which is handy for minutes and invoices, though the model is not perfectly consistent, and large figures spoken quickly are where errors cluster. Check every amount and date against the recording.

Which model should I use for a French interview recorded on a phone?

Accurate or Best. Phone recordings pick up room echo and traffic, and the larger models cope with that far better than Fast. The one-time download is bigger, but the transcript needs much less correction.

Transcribe French Audio to Text — Free & Private, No Sign-up

Convert French audio or video to text in your browser. 100% free, no signup, no upload — ideal for interviews, lectures and meetings.

Model

Spoken

Output

Timestamps [0:05]

⬇️ One-time download: 104 MB (Accurate model 73 MB + 31 MB audio engine, shared with the video tools) — only the first time: it stays saved in your browser, so next time it starts right away. (Translating or switching model needs its own file; a browser can clear this cache if storage runs very low.) Processing takes about 4-8 min per 10 minutes of audio on a typical laptop.

What exactly is downloaded — and is it safe?

• The speech model is OpenAI's Whisper, open source, converted to run in browsers. The audio engine is FFmpeg, the open-source tool behind most video software.
• They come from Hugging Face and unpkg — public repositories used by researchers and developers worldwide, not from us.
• They are not programs. Nothing is installed, no permission is asked, and your browser keeps them in its own cache exactly like it keeps a website's images.
• They can't read your files or touch anything outside this page. Remove them any time by clearing this site's data in your browser.
• Your audio never goes the other way: this site has no server receiving files.

Drop your video or audio here or click to choose

transcribed on your device by OpenAI's Whisper — your recording is not uploaded

ℹ️ Use a recording on your device that you own or have permission to process. This tool cannot fetch YouTube or social video links. Start with a short clip. Larger models can improve accuracy but require more memory and download data. Processing time varies by device, model and recording length. Always review the result.

French punishes transcription tools that work from sound alone. Liaison and elision mean the same syllables can be c'est, s'est, ces or sais depending on what came before. Whisper resolves those from context, which is why its French output reads as written French rather than a phonetic mess, complete with accents, cedillas and sensible punctuation. This page opens with French preselected, so detection cannot wander off to Italian or Catalan on a short clip.

The model runs in your browser, so a recorded entretien, a cours magistral or a réunion is transcribed on your device and never uploaded; there is no minute cap and nothing to pay. Copy the text, download it as TXT, or export SRT or VTT subtitles if the source is a video. For a phone recording in an echoing room or a speaker with a strong accent, choose the Accurate or Best model rather than Fast; the first use downloads the model.

Where it needs a human: proper names, figures, dates and acronyms are the usual errors. Québécois and other regional varieties are handled, with accuracy dipping on heavy slang or very fast colloquial speech. Numbers mostly come out as digits, worth checking in anything financial. Read the draft once against the audio before it goes anywhere official.

More sizes & options

Frequently asked questions

Does it handle Québécois French?: Yes, along with Metropolitan, Belgian and Swiss French. Strong regional slang is where accuracy dips, as it does in any language.
Are accents and punctuation included?: Yes — é, è, ê, ç and proper punctuation are produced automatically, so the transcript reads as written French.
Can I export subtitles?: Yes, as .SRT or .VTT with timestamps, ready for a video editor or YouTube.
Do numbers and dates come out as digits or words?: Mostly as digits, which is handy for minutes and invoices, though the model is not perfectly consistent, and large figures spoken quickly are where errors cluster. Check every amount and date against the recording.
Which model should I use for a French interview recorded on a phone?: Accurate or Best. Phone recordings pick up room echo and traffic, and the larger models cope with that far better than Fast. The one-time download is bigger, but the transcript needs much less correction.

Guides

Related tools

🔒 Private mode (the default): your recording never leaves your device. Instant mode is optional, you switch it on yourself, and it sends the audio to the browser's speech service.