Occitan transcription: be prepared to edit
Occitan is a low-resource language for the speech model, which heard very little of it during training. Expect a rough draft with many corrections. The best results come from slow, clear speech by one person close to the microphone; older speakers in a noisy kitchen or a lively group conversation will need much more work.
A rough draft still helps with long recordings. You listen, click the line that is wrong, fix it, and move on, instead of typing every word yourself.
Upload the recording as it is: MP3, M4A, WAV, OGG, OPUS, FLAC or AAC, a voice note saved from WhatsApp or Telegram, or a video file such as MP4 or MOV. You can also record from your microphone in the browser. Long files are split and processed in parallel, so a ten-minute recording usually takes a minute or two.
Dialects and spelling norms
Occitan is a group of closely related dialects: Provençal, Languedocien, Gascon (including Aranese in the Val d'Aran), Limousin, Auvergnat and Vivaro-Alpine. The model does not separate them well, and a Gascon speaker may come out with forms from another dialect. Correct the text to match the speaker.
Occitan has two main spelling traditions, the classical norm and the Mistralian norm used by many Provençal writers. The transcript does not follow either reliably, so choose the norm you need and apply it during review. Accents such as à, è, ò and ç, and the dot in n·h and s·h, deserve a second look.
Occitan, French and Catalan
Most Occitan speakers also speak French, and recordings switch between the two. Auto-detect often picks French, Catalan or Spanish for an Occitan clip, so this page sets Occitan for you. If a recording is mostly French with Occitan songs or sayings, select French for that file and fix the Occitan lines by hand.
What people record in Occitan
Uploads include interviews with older native speakers for family or local archives, collecting work by associations and researchers, Calandreta school projects, folk song sessions, and radio programmes. Language learners use the draft to check what they heard.
Exports and next steps
Export TXT or DOCX for notes and documents, PDF for a fixed copy, SRT or VTT for subtitles, and CSV with one row per line and its start time. Timestamps and speaker names can be switched on or off in each export. Speaker labels are available on paid plans or with a Minute Pack.
For video, such as a recorded veillée or a YouTube interview, see the Occitan video to text page.