docs / api / voice

POST /v1/voice

Speech-to-text via Whisper. Local-only by default.

POST/api/v1/voice

Transcribe an audio file. Accepts multipart upload.

Parameters

NameTypeRequiredDescription
filebinaryyesAudio file (wav, mp3, m4a, flac, ogg)
languagestringnoISO 639-1 code, e.g. en, nl. Auto-detect if omitted.

Example

curl -X POST http://localhost:8765/api/v1/voice \
-F "file=@audio.wav" \
-F "language=en"