Google AI Music Gets Another Upgrade! Lyria 3.5 Writes Better Songs, Sings with More Nuance, and Offers Precise Control Over Tempo and Duration

Google continues to double down on the generative AI creation track.

On Wednesday, the 29th, US Eastern Time, Google announced the launch of its next-generation AI music generation model, Lyria 3.5, which has been synchronously integrated into Flow Music, the AI music creation tool under Google Labs. Compared to previous versions, the new model focuses on upgrading musicality, lyric generation, AI vocals, and creative control capabilities, enabling users to generate songs that are more natural, structurally complete, and emotionally rich.

This update signifies that, amidst ongoing competition from AI music products like OpenAI, Suno, and Udio, Google is further strengthening its text-to-song generation capabilities. It aims to leverage Flow Music to build a one-stop AI music platform covering creation, editing, and generation.

Four Major Upgrades in Lyria 3.5

According to Google, Lyria 3.5 mainly focuses on upgrades in four areas:

First is Musicality.

Google stated that the new model can generate more complex melodic structures, making the overall music more natural and fluid. The song arrangement also aligns better with real-world music composition logic, rather than simple repetitive loops.

Second is Lyric Generation Capability.

Lyria 3.5 has further improved Prompt Adherence, allowing it to understand user requirements more accurately and generate songs with more reasonable structures and higher-quality lyrics. Improvements have been made in paragraph organization and overall coherence.

Third, and the most significant upgrade this time, is AI Vocals.

Google claims that the vocals generated by the new model offer a more natural and emotionally nuanced singing performance. It has also improved pronunciation accuracy, making the AI singing closer to human performance.

At the same time, users can more precisely control the song tempo and duration, further enhancing creative freedom.

Fourth, Creative Control is Further Enhanced.

In addition to improving generation quality, Google is placing emphasis on "Creative Control" this time.

Google stated that Flow Music aims to make AI more like a creative partner, rather than just a one-off music generation tool.

Users can continuously modify songs using natural language prompts, iterating on melody, lyrics, style, tempo, and overall atmosphere without needing to regenerate the entire song each time.

Google says it hopes to help users "create songs they truly love" through Lyria 3.5, rather than just generating a music demo.

Continuous Iteration of the Lyria Series, Google Doubles Down on AI Music Ecosystem

Lyria is a series of AI music generation models created by Google DeepMind and is an important part of Google's layout in AI content generation in recent years.

In February this year, Google launched Lyria 3; at the end of March, it released Lyria 3 Pro, increasing the maximum song generation duration from about 30 seconds to 3 minutes. It also supports user-defined complete song structures such as Intro, Verse, Chorus, and Bridge, making AI-generated content closer to professional music production workflows.

When releasing Lyria 3.5, Google did not announce new benchmark results, but instead focused the upgrades on user experience aspects such as vocal quality, lyric creation, and creative control. Industry observers believe this means Google hopes to further improve the usability of AI-generated music, rather than purely pursuing model performance metrics.

AI Music Competition Continues to Heat Up

In recent years, AI music has become one of the important directions in the competition for generative AI.

Besides Google, startups like Suno and Udio are rising rapidly, and OpenAI continues to explore multimodal content generation capabilities. Various manufacturers are competing in areas such as song composition, audio editing, and vocal generation.

As an important component of the Google Labs creative tool matrix, Flow Music is building Google's AI content creation ecosystem together with the Flow video generation platform. With the launch of Lyria 3.5, Google hopes to further integrate AI capabilities into the entire music creation process, covering the complete workflow from creative conception and lyric writing to song generation and editing, thereby further enhancing its competitiveness in the field of generative creation.