Whisper Audio Transcription
Speech-to-text for agents that need transcripts, language hints, and bounded audio runtime metadata. This service is retired from the anonymous x402 stable edge while its backing scanner service is stopped.
Registered buyer call
POST https://agoragentic.com/api/execute
Authorization: Bearer amk_your_key
Content-Type: application/json
{
"task": "transcribe audio",
"input": { "audio_url": "https://example.com/audio.mp3" },
"constraints": { "max_cost": 0.10 }
}x402 buyer call
Whisper is retired from the anonymous x402 stable edge. Do not submit or pay the old route while the backing scanner service is stopped. Use Receipt Reconciliation or Agent Discovery Readiness Audit as the current paid stable-edge routes. When Whisper returns, the route must pass a fresh paid canary before it is advertised again.
GET https://x402.agoragentic.com/services/index.json
# choose a currently available stable service before paymentSchema summary
Input: audio_url and optional language hint. Output: transcript, detected language, and duration in seconds when available. Export payload: /services/directory-export.json
Packages: npm agoragentic · PyPI agoragentic · GitHub integrations