Load voice by ID
Return a saved voice's audio, transcript, and metadata by ID.
voice_id*Voice IdSuccessful Response
application/json- response
success*SuccessWhether the voice was loaded successfully
audio_base64?|Base64-encoded reference audio (WAV)
transcript?|Transcript of the reference audio. For design voices this is the natural-language voice description, not the spoken text.
ref_text?|The actual spoken text in the reference audio. Set for design voices; empty string for clone voices.
name?|Voice display name
voice_type?|'clone' or 'design'
has_voice_clone_prompt?|Whether a precomputed voice prompt cache exists
word_timestamps?array<>|Word-level timestamps [{text,start_time,end_time}] if persisted
transcript_manual?|True if the transcript was written or corrected by the user
align_status?|Outcome of the last alignment of this voice's current audio, or null if it has not been aligned
source?|Where the audio was cut from, when it came from a public sample
message?|Error message if load failed
curl -X GET 'http://127.0.0.1:8100/api/v1/voices/string'{ "success": true, "audio_base64": "string", "transcript": "string", "ref_text": "string", "name": "string", "voice_type": "string", "has_voice_clone_prompt": true, "word_timestamps": [ {} ], "transcript_manual": true, "align_status": "done", "source": { "task_id": "string", "start": 0, "end": 1 }, "message": "string"}Save voice to library POST
Save a new voice (audio + transcript) to the voice library. Clone audio can be up to 30 seconds long. A clone longer than 15 seconds is cut to 15 seconds or less, on sentence ends where it can, and the response carries the cut voice.
Delete voice by ID DELETE
Permanently delete a saved voice and its files from disk.