{"slug": "build-a-live-call-transcription-agent-on-telnyx-edge-compute", "title": "Build a Live Call Transcription Agent on Telnyx Edge Compute", "summary": "Telnyx has released an example application that turns live phone calls into actionable data using its Edge Compute platform. The edge-call-transcription-agent answers inbound calls, streams real-time transcription, stores transcript segments in an Agent SDK actor, and uses Telnyx AI Inference to generate a concise call summary that is sent via SMS after hangup. The TypeScript app demonstrates a pattern of combining live call events, durable state, AI summarization, and post-call notifications in a single edge deployment.", "body_md": "Call recordings are useful, but they are usually passive.\n\nThis example turns a live phone call into something your app can use right away.\n\nThe `edge-call-transcription-agent`\n\nexample answers an inbound call on Telnyx Edge Compute, starts streaming transcription, stores final transcript segments in an Agent SDK actor, summarizes the call after hangup with Telnyx AI Inference, persists the record, and sends the summary by SMS.\n\nCode: [https://github.com/team-telnyx/telnyx-code-examples/tree/main/edge-call-transcription-agent](https://github.com/team-telnyx/telnyx-code-examples/tree/main/edge-call-transcription-agent)\n\nThe app is a TypeScript Edge Compute function with two Agent SDK actors:\n\n`TranscribeAgent`\n\n- one actor per call`TranscriptRegistry`\n\n- one shared actor for listing stored transcriptsThe call flow:\n\n``` php\ncall.initiated\n  -> record call state\n  -> answer call\n\ncall.answered\n  -> speak greeting\n\ncall.speak.ended\n  -> start inbound transcription\n\ncall.transcription\n  -> append transcript segment\n\ncall.hangup\n  -> summarize\n  -> store\n  -> SMS summary\n```\n\n`POST /webhooks/voice`\n\nreceives Telnyx Call Control events`GET /`\n\nserves a lightweight dashboard`GET /transcripts`\n\nlists recent stored transcripts`GET /transcripts/:call_control_id`\n\nfetches one transcript`GET /debug/state?call_control_id=...`\n\nshows live actor state`GET /health/liveness`\n\nand `GET /health/readiness`\n\nare health checksStreaming transcription arrives as many events.\n\nThe app needs to remember:\n\nAn actor per call makes that state easy to hold.\n\nThe source maps each `call_control_id`\n\nto one `TranscribeAgent`\n\n, so the transcript can build up across webhook events.\n\nAfter hangup, the actor queues a finalize pipeline:\n\n``` php\nsummarize -> store -> notify\n```\n\nThe summary uses Telnyx AI Inference through the Edge Compute binding:\n\n```\nthis.env.TELNYX.ai.openai.chat.createCompletion({\n  model,\n  messages: [\n    { role: \"system\", content: SUMMARY_SYSTEM_PROMPT },\n    { role: \"user\", content: transcript },\n  ],\n  max_tokens: 200,\n  temperature: 0.3,\n});\n```\n\nThe default model is:\n\n```\nzai-org/GLM-5.2\n```\n\nThe prompt asks for a concise SMS-friendly summary under 320 characters.\n\nOnce the summary is ready, the app sends it over Telnyx Messaging:\n\n```\nthis.env.TELNYX.messages.send({\n  from: this.env.SMS_FROM,\n  to: this.env.SMS_TO,\n  text: state.summary,\n});\n```\n\nUse placeholders in examples and configure real numbers only as private runtime values:\n\n```\nSMS_FROM=<TELNYX_SMS_FROM>\nSMS_TO=<SUMMARY_RECIPIENT>\ngit clone https://github.com/team-telnyx/telnyx-code-examples.git\ncd telnyx-code-examples/edge-call-transcription-agent\nnpm install\n```\n\nSet secrets:\n\n```\ntelnyx-edge secret set TELNYX_API_KEY <YOUR_TELNYX_API_KEY>\ntelnyx-edge secret set SMS_FROM <TELNYX_SMS_FROM>\ntelnyx-edge secret set SMS_TO <SUMMARY_RECIPIENT>\n```\n\nDeploy:\n\n```\ntelnyx-edge ship\n```\n\nPoint your Call Control app webhook to:\n\n```\nhttps://edge-call-transcription-agent-<id>.telnyxcompute.com/webhooks/voice\n```\n\nThen call the Telnyx number assigned to the Call Control application.\n\nBefore using this with real callers, add:\n\nThe useful pattern is simple: live call events, durable state, AI summary, and post-call notification in one edge app.\n\nResources:", "url": "https://wpnews.pro/news/build-a-live-call-transcription-agent-on-telnyx-edge-compute", "canonical_source": "https://dev.to/sonam_50a41a4ced7e6b4f3fa/build-a-live-call-transcription-agent-on-telnyx-edge-compute-1ieo", "published_at": "2026-08-25 05:38:03+00:00", "updated_at": "2026-08-25 06:13:43.198311+00:00", "lang": "en", "topics": ["ai-agents", "natural-language-processing", "generative-ai", "developer-tools"], "entities": ["Telnyx", "Agent SDK", "Telnyx AI Inference", "GLM-5.2", "TranscribeAgent", "TranscriptRegistry", "Telnyx Edge Compute", "Telnyx Messaging"], "alternates": {"html": "https://wpnews.pro/news/build-a-live-call-transcription-agent-on-telnyx-edge-compute", "markdown": "https://wpnews.pro/news/build-a-live-call-transcription-agent-on-telnyx-edge-compute.md", "text": "https://wpnews.pro/news/build-a-live-call-transcription-agent-on-telnyx-edge-compute.txt", "jsonld": "https://wpnews.pro/news/build-a-live-call-transcription-agent-on-telnyx-edge-compute.jsonld"}}