{
  "id": 7660970,
  "title": "Gemini Live audio",
  "url": "https://urgent.news/2026/09/15/gemini-live-audio",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-15T22:47:07.000Z",
  "source": {
    "name": "Simon Willison",
    "slug": "simon-willison",
    "url": "https://simonwillison.net/2026/Sep/15/gemini-live/"
  },
  "original_language": "en",
  "account": "Google has unveiled Gemini 3.8 Live and 3.8 Live Extended Thinking, two new speech-to-speech models inspired by OpenAI's GPT-Live series. To test these models, a custom web user interface was developed, allowing users to choose a model and voice preset, input a system prompt if desired, and engage in a browser-based voice conversation. The interface connects directly to Google's wss://generativelanguage.googleapis.com/ws/google.ai.generativelanguage.v1alpha.GenerativeService.BidiGenerateContent WebSocket endpoint, utilizing the Web Audio API for audio capture and playback. A comprehensive tutorial on utilizing this WebSocket API was released by Simon Willison on September 15th, 2026.",
  "summary": "Tool: Gemini Live audio Google released Gemini 3.8 Live and 3.8 Live Extended Thinking today - two new speech-to-speech models that are a similar shape to OpenAI's GPT-Live family. I pointed GPT-6 Astra Extra High at the documentation and had it build me this web UI for trying out the new models. You can select a model and voice preset, enter an optional system prompt and then start a voice…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "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."
}