whisper audio.mp3 --model small --language es

Category: AI/ML Tooling

whisper audio.mp3 --model small --language es

Transcribe in a specific language

Transcribes Spanish audio using the small model, with --language skipping the auto-detection step and forcing Spanish decoding. Language codes follow ISO 639-1 (es, fr, de, ja, and so on). Forcing the language prevents misdetection on short clips and slightly speeds things up.
Looking for more? Search all 7,657 commands — works offline, in English or Spanish, and fixes typos.