[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-benchmark-measures-whether-ai-tracks-speakers-across-meetings":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},9135,"new-benchmark-measures-whether-ai-tracks-speakers-across-meetings","New Benchmark Measures Whether AI Tracks Speakers Across Meetings","A new evaluation finds an end-to-end system beats five commercial meeting-transcription tools at remembering who said what across separate meetings.","A new benchmark exposes a blind spot in meeting-transcription tools: most lose track of who is speaking from one meeting to the next.\n\nResearchers built a metric called SI-cpWER that scores an entire corpus of meetings against one global speaker-ID assignment, instead of letting each recording relabel anonymous speakers from scratch. They tested five commercial diarize-then-identify cascades, two open academic baselines, and their own end-to-end system, ThyVoice, across the 129-meeting CHiME-8 NOTSOFAR set in clean and noise-augmented form, plus CHiME-6. ThyVoice repairs overlapping speech and filters the evidence used to build and update voiceprints. Across all three conditions, it posted a lower error score than every commercial system tested, and the lowest mean SI-cpWER in the full panel: 47.13, versus 54.75 for the next-best entry.\n\nMost transcription benchmarks still grade each meeting in isolation, so a tool can top the leaderboard while mixing up \"Speaker 3\" between Tuesday's call and Thursday's. That gap matters for anyone building an assistant meant to remember people across months of conversations, not just transcribe a single one. The fact that a research prototype beat five commercial pipelines on this specific test suggests vendors have been optimizing for the wrong number.\n\nCall it the difference between a stenographer and a colleague who actually remembers your name.","[\"ai\",\"speech-transcription\",\"benchmarks\",\"speaker-identification\"]","2026-10-01T04:00:00.000Z","2026-10-01T21:50:49.271Z","2026-10-01T21:50:52.075Z","published",null,[],"ai",[24,26,27,28],"speech-transcription","benchmarks","speaker-identification",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.39344",0,{"sections":35},[36,39,43,47,52,57,61,66,71,75,80,85,90,95],{"name":37,"slug":24,"count":38,"latest_published_at":18},"AI",5572,{"name":40,"slug":41,"count":42,"latest_published_at":18},"Security","security",815,{"name":44,"slug":45,"count":46,"latest_published_at":18},"Policy","policy",430,{"name":48,"slug":49,"count":50,"latest_published_at":51},"Deals","deals",298,"2026-09-30T21:00:26.000Z",{"name":53,"slug":54,"count":55,"latest_published_at":56},"Hardware","hardware",196,"2026-09-30T13:00:00.000Z",{"name":58,"slug":59,"count":60,"latest_published_at":18},"Science","science",163,{"name":62,"slug":63,"count":64,"latest_published_at":65},"Consumer Tech","consumer-tech",149,"2026-09-30T22:57:11.000Z",{"name":67,"slug":68,"count":69,"latest_published_at":70},"Dev Tools","dev-tools",93,"2026-10-01T02:30:48.000Z",{"name":72,"slug":73,"count":69,"latest_published_at":74},"Software","software","2026-09-30T21:41:11.000Z",{"name":76,"slug":77,"count":78,"latest_published_at":79},"Startups","startups",84,"2026-09-30T20:39:09.000Z",{"name":81,"slug":82,"count":83,"latest_published_at":84},"Gaming","gaming",51,"2026-09-30T16:24:30.000Z",{"name":86,"slug":87,"count":88,"latest_published_at":89},"General","general",50,"2026-09-30T21:37:54.000Z",{"name":91,"slug":92,"count":93,"latest_published_at":94},"Reviews","reviews",31,"2026-09-28T14:31:34.000Z",{"name":96,"slug":97,"count":98,"latest_published_at":99},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]