API Reference

POST /v1/voice

Future-ready voice endpoint for speech input and spoken output.

Endpoint reference

Future-ready endpoint

Voice support is planned. This page documents the target request shape so teams can design against it early.

POST/v1/voice

Accept voice or text input and return a voice-ready response payload.

Authentication: Authorization: Bearer YOUR_API_KEY

Request examples

curl https://api.deepconrad.com/v1/voice \
  -H 'Content-Type: application/json' \
  -H 'Authorization: Bearer YOUR_API_KEY' \
  -d '{
    "model": "conrad-5.0",
    "input": {
      "type": "text",
      "content": "Walk me through deploying a Next.js app on Vercel."
    },
    "voice": {
      "format": "pcm16",
      "style": "natural"
    }
  }'

Response example

{
  "id": "voice_01JVOICE",
  "object": "voice.response",
  "model": "conrad-5.0",
  "transcript": "Here is a step-by-step deployment walkthrough...",
  "audio": {
    "format": "pcm16",
    "url": "https://cdn.deepconrad.com/audio/voice_01JVOICE.pcm"
  }
}

Parameters

Parameter	Type	Required	Description
model	string	Yes	Recommended: conrad-5.0 for richer voice workflows.
input	object	Yes	Text or audio input payload.
voice	object	No	Requested output format and style.
conversation_id	string	No	Connects a voice turn to an existing session.

Error handling

Status	Meaning	Notes
400	Invalid voice payload	Check formats and required fields.
401	Unauthorized	Use a valid bearer token.
415	Unsupported media	Use supported audio formats.
429	Rate limited	Retry with backoff.

POST /v1/completions

Build a chatbot