{
  "id": 20121193,
  "title": "Signal2Speech: what a scientist would catch, and letting Claude read the brain pipeline (Update 3)",
  "url": "https://urgent.news/2026/11/15/signal2speech-what-a-scientist-would-catch-and-letting-claude-read",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-15T05:44:50.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/nrupard06/signal2speech-what-a-scientist-would-catch-and-letting-claude-read-the-brain-pipeline-update-3-p89"
  },
  "original_language": "en",
  "account": "Our third update on Signal2Speech, the Hack the Nest 2026 project that enables users to answer questions using an EEG headset, occurs about an hour and a half before midnight. The problem of the headset being lost in transit has been fixed. After grounding was added, Claude Sonnet 5.5 was tested with real AI calls, and the AQI question is now answered correctly using the \"Compare counties\" map option. A new question was answered in two seconds, and the confirmed answer was spoken once. Claude can read the brain pipeline, using a Model Context Protocol (MCP) server to provide live signal pipeline data. The simulator was able to identify a clear P300 response and choose the correct answer three out of three times. The app now shows which pollutant, whether ozone or smoke particles, set the \"before\" AQI value for Washtenaw County. Two adjustments were made to prevent misleading information: a June 26 satellite image now displays cloud cover, and the fire distances were corrected using NASA FIRMS data. The program automatically detects the EEG board and selects the appropriate channels. Future steps include recording the demo video, submitting the project on Devpost, and conducting further testing.",
  "summary": "This is our third update on Signal2Speech, our Hack the Nest 2026 project that lets someone answer questions with an EEG headset. The app suggests replies, the wearer picks one with a P300 response (a brain signal that shows up about 300 milliseconds after the tile you're watching flashes), and they confirm by closing their eyes. Nothing is spoken until they confirm. In our last post, we covered…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "The New Stack",
        "title": "Anthropic now lets you self-host Claude Managed Agents. The catch? 8 hours to act if it flags a safety risk",
        "url": "https://urgent.news/2026/11/13/anthropic-now-lets-you-self-host-claude-managed-agents-the-catch-8",
        "published": "2026-11-13T21:29:30.000Z"
      }
    ]
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}