Audio
OpenAI-compatible speech endpoints. Use your ZeaLLM model group as model.
Base URL: {ZEAGATE_PUBLIC_URL}/v1 with Authorization: Bearer zea-....
Machine-readable HTTP contract: GET /openapi.json on the gateway host.
Speech-to-text
POST /v1/audio/transcriptions
Content-Type: multipart/form-data
| Field | Required | Notes |
|---|---|---|
file | Yes | Audio file |
model | Yes | Model group (e.g. transcribe deployment group) |
stream | No | SSE deltas when true |
Text-to-speech
POST /v1/audio/speech
Content-Type: application/json
| Field | Required | Notes |
|---|---|---|
model | Yes | Model group |
input | Yes | Text to speak |
voice | Yes | Provider voice id |
stream | No | Set true in JSON for streamed audio |
See SDK snippets and the portal API reference page for copy-paste examples.