Create transcription

Transcribe audio files to text, with optional language detection.

Authentication

Authorization Bearer

API key as bearer token in Authorization header.

Request Body

modelstringrequired

Model ID.

languagestring

Input audio language.

promptstring

Transcription prompt.

response_formatstring

Response format.

temperaturenumber

Controls decoding randomness for transcription when supported. Lower values favor stable transcripts; higher values allow more alternatives.

timestamp_granularitiesarray

Timestamp granularity levels.

streamboolean

Streaming response flag.

enable_itnboolean

Converts spoken or formatted entities into normalized written forms, such as numbers, dates, or times, when supported.

enable_puncboolean

Restores punctuation in the transcribed text when supported.

enable_channel_splitboolean

Processes audio channels separately and returns channel-aware transcription results when supported.

enable_ddcboolean

DDC processing flag.

enable_speaker_infoboolean

Returns speaker-related metadata or labels when supported.

corpusobject

Custom corpus configuration.

hotwordsstring

Hotword hints.

request_idstring

Client request ID.

user_idstring

End-user ID.

segment_durationnumber

Audio segment duration.

audio_formatstring

Input audio format.

codecstring

Input audio codec.

rateinteger

Input audio sample rate in Hz.

bitsinteger

Input audio bit depth.

channelinteger

Number of input audio channels.

show_utterancesboolean

Returns utterance-level transcription details, usually including text and timing boundaries, when supported.

enable_nonstreamboolean

Non-streaming recognition flag.

uidstring

Speech recognition user ID.

filestring

Audio file.

file_urlstring

Audio file URL.

audio_datastring

Inline audio data.

metadataobject

Application metadata.

extra_bodyobject

Additional request body fields.

provider_optionsobject

Upstream configuration.

Provide either file or file_url. If both are present, file_url is used. Additional ASR fields are listed in GET /v1/models/{model_id}.parameter_schema.

Response

textstring

Transcribed text.

modelstring

Model ID used.

languagestring

Detected language code. Returned with verbose_json.

durationnumber

Audio duration in seconds. Returned with verbose_json.

taskstring

Transcription task type. Returned with verbose_json.

segmentsobject[]

Timestamped segments. Returned with verbose_json.

segments.idinteger

Segment ID.

segments.startnumber

Segment start time in seconds.

segments.endnumber

Segment end time in seconds.

segments.textstring

Segment text.

utterancesobject[]

Utterance-level details. Returned when supported by the selected model.

utterances.startnumber

Utterance start time in seconds.

utterances.endnumber

Utterance end time in seconds.

utterances.textstring

Utterance text.

utterances.speakerstring

Speaker label when returned.