Create a conversion
Starts a new voice conversion job using either an Audimee built-in voice or one of your custom voice models. Conversions run asynchronously — the response returns immediately with the conversion id; poll GET /conversions/{id} for progress.
When using a custom voice model, the model must have finished training (confirmedDone: true). Calls against models that are still training or that failed training return 400.
Authorizations
Static API key issued by Audimee, passed as Authorization: Bearer {token}.
Body
ID of the voice model to use. Accepts both Audimee built-in IDs (numeric strings) and your own custom model UUIDs.
URL to the audio file to convert. The server downloads it server-side, so the URL must be reachable from the public internet for the duration of the request.
Strength of the voice conversion effect, 0.0–1.0. Defaults to 0.3 when omitted.
0 <= x <= 1Pitch shift applied during conversion, in semitones. Whole numbers between -24 and 24. Defaults to 0.
-24 <= x <= 24Output sample rate. Either 44100 or 48000. Defaults to 44100.
44100, 48000 Response
Conversion job queued. Poll GET /conversions/{id} for progress.
UUID of the queued conversion job.