Skip to navigation
API reference

Transcribe a recording (STT)

Turns a short recording into text.

Send an audio file of up to 60 seconds together with the language spoken in it, and receive the transcript as JSON with the billable duration. Use it for voice notes, call snippets, voice commands and dictation.

Each recording is transcribed in the one language you name, chosen from the Languages page; automatic detection is available only when the catalog lists it. Requests are charged by the duration of the audio, rounded up to the next whole second, at 20 seconds per credit.

Authentication

AuthorizationBearer

Your Llynx API key, sent as Authorization: Bearer <key>.

Request

This endpoint expects a multipart form containing a file.
filefileRequired

The recording: WAV, MP3, M4A, FLAC, Ogg or WebM, mono or stereo, 8–48 kHz, up to 60 seconds and 32 MiB.

languagestringRequired<=63 characters

Language spoken in the recording, as a tag from the Languages page, for example hi or en. auto is accepted only when the catalog lists it.

Response

Transcript and billable duration
idstring

Identifier of this request, also sent as the X-Request-ID header; quote it when you contact us about a charge.

textstring
The transcript.
languagestring
Language used for the transcription.
language_modeenum

explicit when you named the language, auto when it was detected.

Allowed values:
duration_msinteger
Decoded audio duration in milliseconds.
capabilities_revisionstring

Version of the language catalog in force for this request (v1, v2, …), as shown on the Languages page.

usageobject

Errors

400
Bad Request Error
401
Unauthorized Error
402
Payment Required Error
403
Forbidden Error
413
Content Too Large Error
415
Unsupported Media Type Error
422
Unprocessable Entity Error
429
Too Many Requests Error
503
Service Unavailable Error