Talk Webhooks

Webhook events for text-to-speech generation

The Talk API generates events when speech audio is synthesized. These events are delivered through the unified do.dev webhook system.

Event Types

Event Payload

talk.speech.generated

Fired after speech audio is successfully generated.

{
  "eventId": "evt_c3d4e5f6-a7b8-9012-cdef-123456789012",
  "type": "talk.speech.generated",
  "service": "talk",
  "data": {
    "voice": "asteria",
    "format": "mp3",
    "textLength": 256,
    "speed": 1.0
  },
  "livemode": true,
  "createdAt": 1700000000000
}

Subscribing to Talk Events

Create a webhook endpoint that listens to talk events:

curl -X POST https://api.do.dev/v1/webhooks \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://example.com/webhooks",
    "enabled_events": ["talk.*"]
  }'

Use Cases

• Usage monitoring — Track TTS generation volume and character usage across your team
• Cost tracking — Correlate generation events with your usage-based billing
• Content pipeline — Trigger downstream workflows when audio is generated (e.g., upload to CDN, attach to content)
• Voice analytics — Analyze which voices and formats are used most

Related

• Speech API — Text-to-speech endpoint reference
• Webhooks Overview — How the unified event system works
• All Event Types — Complete reference across all services
• Verify Signatures — HMAC-SHA256 signature verification