Google releases Lyria 3.5
Google DeepMind releases the Lyria 3.5 music-generation system. Its model card confirms text prompts as input, music audio and lyric text as output, and distribution through Google Flow Music.
What happened
Event details
Google DeepMind published the Lyria 3.5 model card on July 29, 2026. The system uses latent diffusion over temporal audio latents to generate music from text prompts and can also generate lyric text. Google says it improves audio fidelity and prompt adherence over Lyria 2.
Assessment
Why it matters
Lyria 3.5 advances Google's music generation with higher fidelity and stronger prompt adherence, while documenting clear text input, audio and lyric output, and a product distribution channel.
Availability
Access notes
The official model card is public and lists Google Flow Music as a distribution channel.