Skip to main content
MyVocal exposes several product APIs behind one public host and one accessKey header: Voice, Text-to-Speech (including the Accent model), AI Cover, Text-to-Music and Interpretation.

​
Product lines

​
What you can build with MyVocal

  • Clone a custom voice with the Voice API.
  • Generate speech with that voice using the Text-to-Speech API.
  • Create AI song covers with the same voice using the AI Cover API.
  • Generate original music from a text brief with the Text-to-Music API.
  • Translate and dub an uploaded audio or video file with the Interpretation API.

​
Core workflow

Voice-based integrations usually follow a simple flow:
  1. Clone or prepare a voice
  2. Use the voice for text-to-speech
  3. Use the same voice for AI song covers
Text-to-Music and Interpretation are not voice-dependent. They take a text brief (Text-to-Music) or an uploaded media file (Interpretation), so they do not require a cloned voice, and they do not require a workspace/browser session. Both authenticate with the same accessKey header.

​
Synchronous, streaming and asynchronous APIs

  • Voice and user info return JSON immediately.
  • Text-to-Speech returns an audio stream; read Content-Type before trusting the body.
  • AI Cover uses a callback URL that you provide. See Callback.
  • Text-to-Music and Interpretation are poll-based. They return a project/job identifier immediately, then you poll the project or job endpoints until the resource reaches a terminal state. They do not require a callback URL and do not deliver webhooks.

​
How TTS models work

If you are calling the Text-to-Speech API, you can either use the default model path or explicitly choose a public modelId such as myvocal_v3.
  • Default model path: do not send modelId.
  • Explicit model path: send a public modelId, for example myvocal_v3.
Publicly documented model IDs in this phase:
English(US) currently uses the default model path. A public explicit modelId for English(US) is not published in this phase.

​
Next steps

Quickstart

Make your first MyVocal API call in a few steps.

Authentication

Use your accessKey to authenticate API requests.

TTS Models & Migration

Learn when to use the default TTS path and when to use myvocal_v3.

Text-to-Music quickstart

Go from a creative brief to a downloadable MP3.

Interpretation quickstart

Upload a file and produce translated dubbing.

Characters, quotes and retries

Understand quoting, reservation, settlement and release.