Google's most advanced AI music generation model arrives in the Gemini app, with full availability for developers via the API in Google AI Studio.
اضافة اعلان
Google has added "Lyria 3.5," its latest and most advanced AI music generation model, to the Gemini app. The model was previously available through Google's AI filmmaking tool "Flow," before the company began rolling it out to all Gemini users.
This move lets users create songs, musical pieces, and soundtracks with higher quality, relying on simple text prompts or even using images as a starting point.
More Natural AI-Generated Music
Google says "Lyria 3.5" delivers richer musical arrangements and more expressive vocal performance, alongside improved understanding of user prompts compared to previous versions.
The model can create complete songs including verses, choruses, and transitional sections, rather than simply repeating a short melodic loop as was the case with some earlier music generation tools.
From within the Gemini app, users can now choose a music genre or describe it in words, specify whether they want a vocal piece or instrumental music, and choose between a short clip or a longer musical work.
Beginner users can rely on ready-made templates to start the creation process, whether to produce background music for a video, a custom promotional jingle, or a personal ringtone.
However, the model imposes an important limitation: users cannot edit the musical piece during the generation process once it has been created through the same prompt.
At the same time, Google is making "Lyria 3.5" available through the Gemini API, giving developers the ability to integrate AI music generation technology into their own applications and services.
Lyria 3.5 Enters a Turbulent Music Market
The model's launch comes at a time when music platforms are seeing an increasing flood of AI-generated content.
On the Deezer platform, for example, AI-generated songs now account for about 44% of all musical works uploaded daily, while it later emerged that a significant proportion of the listens some of these works received were fraudulent.
For this reason, Google embeds an invisible digital watermark using "SynthID" technology in every track generated by "Lyria," aimed at enabling identification of AI-produced content.
Lyria's safeguards also include restrictions preventing the cloning of a specific artist's voice or the reproduction of copyrighted song lyrics.
These measures come amid ongoing legal debate over AI-generated music, particularly following a recent European court ruling against competing AI music generation company Suno, related to copyright issues.
However, the ruling does not mark the end of the legal dispute, as it remains subject to appeal, meaning the legal status of AI-produced music remains unresolved for now.
