Generate music from a text prompt.
Providers that do not support music generation should throw an error
and have supports return false for 'music'.
The music generation parameters.
A result envelope containing one or more generated audio clips.
Optional generateSFXGenerate a sound effect from a text prompt.
This method is optional — providers that do not support SFX generation
should either omit it or have supports return false for
'sfx'.
The SFX generation parameters.
A result envelope containing one or more generated audio clips.
Optional shutdownReadonly providerUnique identifier for this provider (e.g. 'suno', 'elevenlabs-sfx').
Readonly isWhether initialize has been called successfully.
Optional Readonly defaultDefault model used when the request omits modelId.
Abstraction over an audio generation backend (Suno, Udio, Stable Audio, ElevenLabs, Replicate, etc.).
Capability negotiation
Not every provider supports every sub-modality. The supports method lets callers (and the FallbackAudioProxy) query whether a given capability is available before invoking it.
Lifecycle
Example