HopBase DocsUpload clone audio
Cloning flow step 1: upload a recording or example clip to get a single-use `file_id`; free.
/v1/files/uploadSubmitted as multipart/form-data with only the purpose and file fields; any other field returns 400. voice_clone is the recording to clone, and prompt_audio is an optional example clip. Format and size are checked at upload; duration (voice_clone 10 seconds to 5 minutes, prompt_audio under 8 seconds) is checked when you clone.
Files are single-use: a successful clone automatically deletes both the source audio and the example audio; a failed clone keeps them so you can retry with other parameters. Files not used within 24 hours of upload are deleted automatically. Uploaded audio cannot be downloaded. The key's group must include the MiniMax Speech models.
Clone only your own voice, or one whose speaker has given you explicit written permission.
Headers
Bearer sk-…: an API key created in the console under API keys; its group must include the requested model
Body parametersmultipart
voice_clone for the recording to clone, prompt_audio for an example clip
Valuesvoice_cloneprompt_audio
One file with extension .mp3 / .m4a / .wav, ≤20 MB
Returns
200Uploaded
Uploaded file
File ID; pass it as file_id or clone_prompt.prompt_audio when cloning
File size in bytes
Upload time, Unix seconds
File name
voice_clone or prompt_audio
Valuesvoice_cloneprompt_audio
{"status_code": 0, "status_msg": "success"} on success; errors do not come here but as a non-200 status code with an error object
0 on success
success on success
Errors
purpose, empty file, unsupported format, or an extra file or fieldmissing_api_key / invalid_api_key / api_key_expired)model_not_found), or the path does not belong to the group (route_not_found)file exceeds the 20 MiB limit