{
  "$type": "site.standard.document",
  "bskyPostRef": {
    "cid": "bafyreica4xpefdmexk2eistgis4vswdb7l54kkm46bgrfkelggcegtozsu",
    "uri": "at://did:plc:4jjxx3max7tcdxwmdkjrnyj4/app.bsky.feed.post/3mfefznowp362"
  },
  "coverImage": {
    "$type": "blob",
    "ref": {
      "$link": "bafkreiacbedqffr4kthbyagevcuz2dm6347nhulu2h7cfemmgbz6aesfs4"
    },
    "mimeType": "image/jpeg",
    "size": 77433
  },
  "path": "/google-lyria-3-ai-music-generation-model",
  "publishedAt": "2026-02-20T17:10:48.000Z",
  "site": "https://nofilmschool.com",
  "tags": [
    "Google",
    "Google gemini",
    "Google deepmind",
    "Google ai",
    "Lyria 3",
    "Ai music",
    "Ai music generator",
    "Ai soundtrack",
    "Google lyria 3 ai music generation model",
    "_Seedance 2.0_",
    "_Kling 3.0_",
    "_Google’s Veo 3_",
    "_Google on the launch page here_",
    "_SynthID_",
    "_company’s website here_"
  ],
  "textContent": "\n\n\n\nAs we’ve seen with the most recent AI video models like _Seedance 2.0_, _Kling 3.0_, and _Google’s Veo 3_, the AI wars are heating up once again. And while the debates surrounding the capabilities, ethics, and actual usability of the AI video models continue to rage on, the other non-video AI breakthroughs also keep coming out as well.\n\nIn the realm of audio, music, and soundtracks, Google DeepMind has released its latest AI music generation model with Lyria 3.\n\nThis new music-gen model promises to be the company’s most advanced version yet, as it promises to be able to create custom soundtracks in seconds with a focus on 30-second clips, auto-generated lyrics, custom cover art, and even somewhat of a focus on ethics and accountability too.\n\n* * *\n\n### Google’s Lyria 3 AI Music Generation Model\n\n\n\n\nAnnounced as a new option within the Google Gemini app, users will now have the ability to make 30-second tracks using text or image prompts. It works as you’d imagine, simply describe your vision or upload from your camera roll, and watch a track come to life in seconds.\n\nRolling out as a beta version first, Lyria 3 works improves upon the audio generation from previous Lyria models in three main ways, as outlined by _Google on the launch page here_.\n\n  1. No need to provide your own lyrics! They'll be generated for you based on your prompt.\n  2. You have more creative control over elements like the style, vocals, and tempo you want.\n  3. You can create more realistic and musically complex tracks.\n\n\n\n### How to Try Out Lyria 3\n\n\n\n\nAgain, no one is forcing you to adopt or even try out these new AI models and tools. There’s also the likely chance that, even with these improvements, AI-generated soundtracks aren’t going to sound as good as the real thing.\n\nIf you are curious about Lyria 3, though, here’s how you can use it:\n\n  * Text to track: Describe a specific genre, mood, inside joke, or memory to create unique tracks with lyrics or instrumental audio that fits your vibe. “I’m feeling nostalgic. Create a track for my mother about the great times we had as kids and the memories of her home-cooked plantains. Make it a fun afrobeat track with a true African vibe.”\n  * From photos and videos to track: Upload a photo or video and watch Gemini use the content to compose a track with lyrics that fit the mood perfectly. “Use these photos to create a track about my dog Duncan on a hike in the woods.”\n\n\n\n### SynthID and AI Responsibility\n\n\n\n\nWith the legality and ethics of AI being, perhaps, by far just as important as any news of new models or technologies these days, it’s is important to note that Google has shared that all tracks generated in the Gemini app (and with this Lyria 3 model) will be embedded with _SynthID_, the company’s imperceptible watermark for identifying Google AI-generated content.\n\nGoogle is also offering the ability for anyone looking to verify if any audio is AI-generated by allowing users the ability to upload any file and simply ask if it was generated using Google AI, and Gemini will check for SynthID and use its own reasoning to return a response.\n\nWe also should mention that Google has shared a bit more about its aims to develop generative AI responsibility and has been “very mindful of copyright and partner agreements as we've trained Lyria 3.”\n\nYou can find out more about Lyria 3 and Google Gemini’s latest news, updates, and promises of responsibility on the _company’s website here_.",
  "title": "Google Launches Lyria 3: Exploring the Capabilities and Ethics of This New AI Music Generator"
}