{"uid":"cap_hOcB47qKmLg9VfSHYkVZ0","slug":"cnvrt-live-stream-analysis-api-8e4e2aa0","name":"CNVRT Live Stream Perception API","description":"Real-time analysis of live streams (HLS/DASH/RTMP/SRT) with ASR, vision, and embeddings. Supports URL-based streams and agent chunked upload. Agent-only endpoint. Payment via USDC on base.","url":"https://cnvrt.ing/api/convertlive","method":"POST","headers":{},"bodySchema":null,"responseSchema":{"input":{"type":"http","method":"POST","bodyType":"json","bodyFields":{"url":{"type":"string","required":false,"description":"Live stream URL (HLS, DASH, RTMP, SRT, WebRTC, Twitch, TikTok, YouTube). VOD/static URLs will be rejected. Not required if source_type is 'agent_capture'."},"webhook":{"type":"string","required":false,"description":"Webhook URL to receive events (optional)"},"features":{"type":"array","required":false,"description":"Features to enable: asr, vision, ocr, embeddings. Default: [asr, vision]"},"source_type":{"type":"string","required":false,"description":"Source type: 'url' (default) or 'agent_capture'. Use 'agent_capture' for zero-friction chunked upload from agents."},"session_meta":{"type":"object","required":false,"description":"Optional metadata for the session"},"latency_target_ms":{"type":"number","required":false,"description":"Target latency in milliseconds. Default: 1000"}}},"output":{"type":"object","required":["stream_id","events_ws"],"properties":{"status":{"type":"string","description":"Session status"},"events_ws":{"type":"string","description":"WebSocket URL for real-time events"},"stream_id":{"type":"string","description":"Unique stream session identifier"},"upload_url":{"type":"string","description":"URL to POST media chunks (agent_capture mode only)"}}}},"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.05","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.05/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.05","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.05","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_9WYeSHoDzVIGbC_VBK08I","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.05","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Real-time analysis of live video/audio streams (HLS/DASH/RTMP/SRT) with ASR transcription, vision, OCR, and embeddings via a WebSocket event stream","exampleAgentPrompt":"Start analyzing this live HLS stream at https://example.com/live.m3u8 in real time — enable ASR and vision features, target under 1000ms latency, and send events to my webhook at https://myapp.io/hook.","exampleUseCases":null,"resultDescription":"Returns a stream_id, a WebSocket URL (events_ws) to receive real-time analysis events (transcripts, detections, OCR, embeddings), and — if using agent_capture mode — an upload_url to POST media chunks to. Events arrive continuously over WebSocket as the stream is processed.","failureModes":["VOD or static URL submitted instead of a live stream — request rejected with error","Unsupported stream protocol or unreachable stream URL — session fails to initialize","Invalid feature names in features array — unknown features ignored or error returned","Agent_capture mode without using upload_url — no data received, session times out","WebSocket connection dropped — real-time events lost, reconnection required","Payment failure or insufficient USDC balance — 402 response, no session created"],"whenToPreferThis":"Choose this endpoint when you need real-time, low-latency analysis of an actively live stream (HLS, DASH, RTMP, SRT, Twitch, TikTok, YouTube Live) rather than a static file. It is ideal when you need multiple simultaneous perception modalities (ASR + vision + OCR + embeddings) in a single session with WebSocket event delivery. For pre-recorded or static file transcription, use CNVRT's file upload or URL transcription endpoints instead.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-09-14T12:38:32.656Z","isFirstParty":false}