Google rolls out Lyria 3.5 music model in Gemini with vocals and longer tracks
cedric_chee · x · 2026-09-05
Google has introduced Lyria 3.5 in the Gemini app, calling it its most advanced music generation model with richer arrangements, more expressive vocals, and higher fidelity. Users can pick a genre, choose between vocal or instrumental output, explore new templates, and generate both short and longer tracks. The poster notes it has been available in AI Studio since the previous day.
More from Multimodal
- Suno V6 music generation model is about to launch — op7418 · 2026-09-05
- pyannote speaker-diarization-community-1 trends on Hugging Face — pyannote · 2026-09-05
- Creator builds an interactive 90s sitcom genie using fal's H3 Max video model — gorkem · 2026-09-05
- Orbis paper reframes video generation as a steerable, continuous visual process — rohanpaul_ai · 2026-09-05
- Pocket FM killed its business model at $200M ARR to bet everything on AI content — thisguyknowsai · 2026-09-05
- Intangible ships Motion Paths: draw a line in 3D and subjects follow it exactly — azed_ai · 2026-09-05