Google Unveils Lyria 3.5 for More Natural AI Music
Google has launched Lyria 3.5. This is the latest version of its AI music generation model. The update focuses on making AI-created songs sound more natural and expressive.
The new model improves several key areas. It creates more detailed melodies. It also produces lyrics that stay closer to user prompts. In addition, the lyrics follow a more consistent song structure. Vocals sound more expressive now. Pronunciation has improved as well.
Users gain greater control too. They can adjust the tempo and duration of tracks more easily. Lyria 3.5 is rolling out today through Google Flow Music.
This release builds on Google’s earlier Lyria models. It continues the company’s work to improve AI-powered music tools.
Better Creative Tools in Flow Music
Flow Music now offers more than basic song generation. Users can already edit individual parts of a track. They can rewrite or translate lyrics. They can also change instruments and refine the music. There is no need to recreate the entire composition.
At Google I/O 2026, the company showed new conversational editing tools. These tools help with music videos. They also give more precise control over individual elements in a track.
Lyria’s Journey So Far
Google first introduced Lyria in 2023. It came from Google DeepMind as a music generation model. Earlier this year, the company launched Lyria 3. That version supported songs from text prompts, images, and videos. It could generate lyrics, vocals, and cover art. Users could adjust genre, mood, and tempo.
Later, Google released Lyria 3 Pro. It extended song length to three minutes. It also added more creative controls for paid Gemini subscribers. Lyria has since reached YouTube Shorts through Dream Track. It is also available in Google Vids for AI-generated background music.
Google has not confirmed whether Lyria 3.5 will fully replace Lyria 3 Pro. Access depends on the subscription. Google AI Plus, AI Pro, and AI Ultra users get different Flow Music tiers.