[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-method-aligns-brain-signals-to-both-audio-and-text":10,"sections":35},{"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":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},8054,"new-method-aligns-brain-signals-to-both-audio-and-text","New Method Aligns Brain Signals to Both Audio and Text","A new brain-decoding framework aligns neural signals to audio and text together, fixing a tradeoff that hurt earlier single-anchor approaches.","A brain-computer interface model now reads neural signals by matching them to audio and text at the same time, not just one or the other.\n\nResearchers built BAT-CLIP, a CLIP-style alignment framework for intracranial EEG (iEEG) recordings that jointly aligns neural embeddings to both audio and text anchors inside a shared, frozen audio-text embedding space. Earlier brain-speech decoding systems anchored neural activity to a single modality: audio, which preserves timing but blurs meaning, or text, which captures meaning but loses acoustic detail. BAT-CLIP splits that difference by training against both anchors at once. Tested on the naturalistic Podcast benchmark, it produced more robust neural representations than bimodal baselines that anchor to only one modality.\n\nThe result matters because it mirrors a broader shift in brain decoding research: borrowing multimodal alignment tricks from mainstream AI, like CLIP's image-text pairing, and retrofitting them for neural signals instead of pixels. The researchers also found that the choice of anchor model matters - using self-supervised foundation models to generate the audio-text targets produced better results, a detail that will shape how future brain-interface labs pick their pretrained building blocks.\n\nThe paper doesn't report the kind of decoding accuracy numbers that would show this closes the gap keeping speech-generating brain interfaces mostly in the lab. It's a methods improvement, not a working product.","[\"brain-computer-interface\",\"neuroscience\",\"multimodal-ai\",\"machine-learning\"]","2026-09-28T04:00:00.000Z","2026-09-28T07:35:50.602Z","2026-09-28T07:35:57.698Z","published",null,[],"ai",[26,27,28,29],"brain-computer-interface","neuroscience","multimodal-ai","machine-learning",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.31180",0,{"sections":36},[37,40,44,49,54,59,63,68,73,78,83,88,92,97],{"name":38,"slug":24,"count":39,"latest_published_at":18},"AI",4750,{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",759,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",399,"2026-09-27T18:39:02.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Deals","deals",261,"2026-09-27T15:30:35.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":58},"Hardware","hardware",188,"2026-09-27T20:46:36.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":18},"Science","science",151,{"name":64,"slug":65,"count":66,"latest_published_at":67},"Consumer Tech","consumer-tech",135,"2026-09-26T14:30:00.000Z",{"name":69,"slug":70,"count":71,"latest_published_at":72},"Software","software",91,"2026-09-25T20:55:00.000Z",{"name":74,"slug":75,"count":76,"latest_published_at":77},"Dev Tools","dev-tools",84,"2026-09-26T04:20:58.000Z",{"name":79,"slug":80,"count":81,"latest_published_at":82},"Startups","startups",76,"2026-09-25T18:33:59.000Z",{"name":84,"slug":85,"count":86,"latest_published_at":87},"Gaming","gaming",48,"2026-09-25T18:35:21.000Z",{"name":89,"slug":90,"count":86,"latest_published_at":91},"General","general","2026-09-26T17:02:42.000Z",{"name":93,"slug":94,"count":95,"latest_published_at":96},"Reviews","reviews",30,"2026-09-24T20:07:31.000Z",{"name":98,"slug":99,"count":100,"latest_published_at":101},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]