[{"data":1,"prerenderedAt":4},["ShallowReactive",2],{"article-doc:docs\u002Frecorded\u002Fstructured-results":3},"---\ntitle: Structured Results\ndescription: Recorded Speech can return structured outputs alongside the transcript.\nstatus: stable\n---\n\n## Segment list\n\nEach segment's text, timestamps, and confidence. The core structured output.\n\n```json [Response]\n{\n  \"id\": 0,\n  \"start_time\": 0.0,\n  \"end_time\": 3.2,\n  \"text\": \"The weather is nice today\",\n  \"confidence\": 0.95\n}\n```\n\n## Summary statistics\n\nAudio-level statistics:\n\n```json [Response]\n{\n  \"summary\": {\n    \"total_duration\": 120.5,\n    \"total_speech_duration\": 95.3,\n    \"overall_speech_ratio\": 0.79,\n    \"num_segments\": 45\n  }\n}\n```\n\n## Metadata\n\nProcessing metadata:\n\n```json [Response]\n{\n  \"metadata\": {\n    \"language\": \"zh\",\n    \"model\": \"whisper-large-v3-turbo\",\n    \"chunk_count\": 3,\n    \"audio_duration_seconds\": 120.5\n  }\n}\n```\n\n## LLM review results\n\nWhen `review: true` is enabled, the reviewed transcript replaces the raw result:\n\n1. **Per-chunk review**: each chunk independently corrected\n2. **Global review**: final correction after aggregation\n\nReviewed results are stored in `reviewed_result.json` in R2.\n\n## Not yet stable\n\n| Capability | Status |\n|---|---|\n| Automatic summary (text) | Roadmap |\n| Chapter segmentation | Roadmap |\n| Action item extraction | Roadmap |\n| Speaker diarization | Roadmap |\n\n::callout{icon=\"i-lucide-triangle-alert\" color=\"amber\"}\nDo not treat experimental or unmarked capabilities as stable. If a capability is marked as \"Roadmap\", it is not yet implemented or may change.\n::\n\n## Stability markers\n\n| Marker | Meaning |\n|---|---|\n| Stable | Shipped. Safe to depend on in production. |\n| Preview | Available, but contracts may still change. |\n| Experimental | May change or be removed without notice. |\n| Roadmap | Planned, not yet implemented. |\n\n## Related\n\n- [Transcribe Audio](\u002Fdocs\u002Frecorded\u002Ftranscribe-audio) — get results\n- [Timestamps & Speakers](\u002Fdocs\u002Frecorded\u002Ftimestamps-speakers) — timestamp details\n- [Batch Jobs API](\u002Fdocs\u002Fapi-reference\u002Fbatch-jobs) — API reference\n",1790059118961]