Audio Models
Text-to-speech, transcription, and music/audio generation through OpenAI-compatible audio routes and multimodal chat.
Example audio models
The examples below may change. Use GET /v1/models to confirm which audio-capable models are available to your key.
| Parameter | Type | Required | Description |
|---|---|---|---|
| openai/gpt-audio | OpenAI | optional | Speech generation and audio-capable chat. |
| openai/gpt-audio-mini | OpenAI | optional | Compact audio-capable chat tier. |
| google/lyria-3-pro-preview | optional | Music generation (Lyria). | |
| google/lyria-3-clip-preview | optional | Short music clip generation. |
Common tasks
| Parameter | Type | Required | Description |
|---|---|---|---|
| Text-to-speech | task | optional | Synthesize speech via `/v1/audio/speech`. |
| Transcription | task | optional | Convert audio to text via `/v1/audio/transcriptions`. |
| Audio-in chat | task | optional | Send `input_audio` parts to audio-capable chat models. |
| Music generation | task | optional | Generate music with Lyria-class models. |
Examples
See the Audio API page for full request bodies and runnable text-to-speech and transcription examples.
Model-dependent features
OneRouter keeps protocol compatibility and forwards model-specific parameters where supported. Features such as tool calling, structured output, reasoning, prompt caching, web search, image input, audio input, video generation, and provider-specific options depend on the selected model and upstream provider. Check the model detail page before relying on a feature in production.