Spanish audio and video to text

Regional variety

Spanish is spoken across many countries with different accents, vocabulary and speed of speech. Results vary between them, and a recording with a clear microphone helps more than the accent does. Expect the occasional wrong word for local slang.

Accents and punctuation

Check the accents on words (á, é, ñ) and the opening question and exclamation marks when you proof-read: the editor lets you correct any line. Names of people and places are the commonest corrections.

What people use it for

Recorded classes, interviews, family voice messages, meetings with Spanish-speaking colleagues and videos that need Spanish captions.

Which model works best for Spanish

Spanish is one of the languages Whisper-family models have seen the most speech in, so the small model can give a reasonable first draft from clear audio. That is not a promise of accuracy: it still slips on names, numbers, jargon and overlapping or noisy speech, so proof-read anything you will publish or rely on.

On tryscriptorium.com the no-account preview and regular accounts run the small model (the larger medium and large-v3 models run when you self-host Scriptorium).

Spanish is among the languages with the most training speech, and the small model takes both Latin American and European speech. Spanglish and English names inside Spanish sentences are the lines to check first.

Subtitles and captions in Spanish

Spanish runs longer than English, so keep lines near 38 to 42 characters and let a subtitle split before a conjunction. The opening ¿ and ¡ are plain Unicode and survive in SRT and VTT.

The uploader is set to Spanish

The upload box on this page transcribes as Spanish, so the recogniser does not have to guess the language from the first seconds. If the recording is in another language or switches languages, use Audio to text, which detects the language automatically. Automatic transcription is a draft and no accuracy figure is promised: check names, numbers and anything you will rely on.