Basque audio to text: what to expect
Basque is a mid-resource language for the speech model. Clear speech gives a usable draft, but you should expect to read it carefully. Long agglutinated words are where it slips: a case ending or verb auxiliary can come out slightly wrong even when the root is right. Names and place names need checking too, and so do numbers, which Basque counts in twenties (hogei, berrogei) and which are easy to mishear in fast speech.
Upload the recording as it is: MP3, M4A, WAV, OGG, OPUS, FLAC, a voice message or a phone recording. You can also record from your microphone in the upload box. Nothing needs converting first.
The transcript opens next to a player that highlights the current line. Click a line to hear it again, correct the word in place, and your changes save automatically.
Batua and the dialects
The model writes closest to euskara batua, the standard used in schools, media and most written Basque. Speakers of Bizkaiera, Gipuzkera, Nafarrera or the Zuberera of Iparralde use different vocabulary, verb forms and pronunciation. Their speech is transcribed, but the text tends to be pulled toward batua spellings or contains more errors, so dialect recordings need a closer review. If you are transcribing for a dialect archive, keep the speaker's forms when you correct the text rather than letting batua spellings stand.
Spelling features such as tx, tz, ts, x and the ñ in words like andereño are kept. Exports are UTF-8 throughout. Speakers in Iparralde also pronounce some sounds differently from speakers in Gipuzkoa or Bizkaia, which can change how a word is spelled in the draft.
Spanish and French mixed in
Almost every Basque speaker is bilingual. South of the border speech mixes in Spanish words and phrases; in Iparralde it mixes in French. Short borrowed words usually come through. Choose the language by what dominates the recording:
- Mostly Basque with Spanish or French phrases: keep Basque selected.
- A Spanish-language meeting with some Basque: choose Spanish.
- Auto-detect judges the opening seconds and can pick Spanish for a Basque file that starts with a Spanish greeting, so set the language yourself.
Who transcribes Basque and what next
Typical recordings are university lectures and ikastola materials, research and oral-history interviews, local radio and podcasts, council meetings and bertsolaritza sessions. Improvised sung verse is hard for any speech model, so bertso recordings need heavy editing.
Export TXT, DOCX, PDF, CSV, SRT or VTT, with timestamps and speaker names switched on or off per export. Speaker labels are available on paid plans or with a Minute Pack.
For video files, see the Basque video to text page.