GET

/v1/schemas/cartesia/audio/stt?kind=output

Accept:

kind outputactivity audiohash bc2b257a…extracted Aug 15, 2026version currentsource upstream spec ↗

field	recorded source
schema	Source linked above. upstream-spec Source trace `6739c32aeeb12ccbed66a4247491fb9346ebc69c6c52bbbd29d25334423d09c8` { "fetchedAt": 1786770738, "extractorVersion": "1", "verifiedAt": null }

Response body

7 properties · 2 required · 1 $defs

property	type	description	constraints
type*	enum	The message type. Always `transcript` for a batch transcription response.	transcript
text*	string	The transcribed text.
request_id	string	Unique identifier for this transcription request.
is_final	boolean	Not used for batch transcription.	deprecated
language	string	The specified language of the input audio.
duration	number	The duration of the input audio in seconds.	format: double
words	TranscriptionWord[]	Word-level timestamps showing the start and end time of each word. Only included when `[word]` is passed into `timestamp_granularities[]`.

Types

1 $defs

TranscriptionWordobject

word*	string	The transcribed word.
start*	number	Start time of the word in seconds.	format: double
end*	number	End time of the word in seconds.	format: double

Validate a payload

before you spend tokens

shell

$ curl -X POST https://modelschemas.com/v1/validate \
    -d '{"provider":"cartesia","endpointId":"stt","payload":{…}}'
→ {"valid": false, "errors": [{"path": "#", "keyword": "required", …}]}