multipart/form-data

Upload

The audio file is a part of the same request that creates the job.

Field	Type	Description
`model`Required	string	Catalogue id. Send this part before the file. A missing model is rejected with model_required.
`file`Required	file	Audio file for that product.
`params`	string	JSON object for that model, encoded as a string. Omit the field to use the defaults. The fields are on the model page.
`reference`	file	Second clip on POST /v1/voice-conversion. This voice replaces the source voice.
`reference_text`	string	Transcript of the reference clip. Required on voice continuation, and on the voice-cloning models that ask for it.
`name`	string	Optional label stored on the job and returned as name.

Size

Speech to text, stem separation, voice isolator, and audio enhancement accept a file under 500 MB. Pascal Transcribe accepts a file under 80 MB. Sarvam Saaras accepts audio up to 2 hours. Sarvam Dubbing accepts media up to 1 hour.

Voice cloning, voice continuation, and voice conversion accept a file under 25 MB.