[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"article-doc:docs\u002Frecorded\u002F1.index":3},"---\ntitle: Recorded Overview\ndescription: Batch transcription jobs from audio URLs.\nstatus: stable\n---\n\n# Recorded Overview\n\nRecorded is **batch transcription**: submit a public `audio_url`, process asynchronously, receive complete results.\n\nFor **synchronous client-VAD clips** see [Segment Transcription](\u002Fdocs\u002Fapi-reference\u002Fsegment-transcription). For **live streams** see [Realtime](\u002Fdocs\u002Frealtime).\n\n## How it works\n\n```\nSubmit audio URL (POST \u002Fv1\u002Faudio\u002Ftranscriptions\u002Fjobs)\n  → Probe audio format and duration\n  → Slice into time-based chunks\n  → Concurrent transcription\n  → Aggregate (merge into global timestamps)\n  → Optional LLM review\n  → Webhook callback (optional)\n```\n\n## When to use\n\n| Scenario | Suitable |\n|---|---|\n| Post-meeting processing | ✅ |\n| Batch audio file transcription | ✅ |\n| Need precise timestamps | ✅ |\n| Need structured results | ✅ |\n| Need subtitle files | ✅ |\n| Short utterance after client VAD | ❌ use [Segment](\u002Fdocs\u002Fapi-reference\u002Fsegment-transcription) |\n| Live microphone stream | ❌ use [Realtime](\u002Fdocs\u002Frealtime) |\n\n## vs. Realtime and Segment\n\n| Dimension | Batch jobs | Segment HTTP | Realtime |\n|---|---|---|---|\n| Transport | HTTP async | HTTP sync | WebSocket |\n| Input | Audio URL | Multipart `file` | PCM16LE stream |\n| Latency | Minutes | ~≤1.6s | Milliseconds |\n| Output | Complete JSON | Transcript JSON | Event stream |\n| Segmentation | **Server** (time slice) | **Client** (VAD) | **Server** (VAD) |\n| Success | **202** + poll | **200** | WS events |\n\n## Current capabilities\n\n| Capability | Status |\n|---|---|\n| Full-file transcription | ✅ |\n| Segment timestamps | ✅ |\n| Confidence scores | ✅ |\n| Translation | Coming soon |\n| Webhook callback | ✅ |\n| LLM review correction | ✅ (optional) |\n| Speaker diarization | Coming soon |\n| SRT\u002FVTT subtitles | ✅ (generated from timestamps) |\n\n## Next steps\n\n- [Transcribe Audio](\u002Fdocs\u002Frecorded\u002Ftranscribe-audio) — submit a job\n- [Batch Jobs API](\u002Fdocs\u002Fapi-reference\u002Fbatch-jobs)\n- [Segment API](\u002Fdocs\u002Fapi-reference\u002Fsegment-transcription) — client-VAD clips\n- [Timestamps & Speakers](\u002Fdocs\u002Frecorded\u002Ftimestamps-speakers)\n",1790059118959]