news 
Google’s enterprise cloud gets a music-generating AI model
Today, Google announced updates to its first-party media-generating AI models available through its Vertex AI cloud platform. This includes the release of Lyria, a text-to-music model that is now in preview for select customers.
Lyria allows users to create songs in various styles and genres, from jazzy piano solos to lo-fi tracks, by providing an alternative to royalty-free music libraries. The company highlights the potential benefits of using its AI-generated music in corporate environments, such as creating branded content or enhancing user experiences.
Chirp 3, another AI model, can synthesize speech in around 35 languages and is now generally available. This model powers Instant Custom Voice, which can clone a voice with just 10 seconds of audio. Additionally, a new tool called Transcription with Diarization has been launched in preview, capable of separating and identifying speakers in recordings with multiple participants.
The company has also enhanced its Veo 2 video creation model, offering the ability to remove background images, logos, and objects from existing videos, extend the frame of video footage (e.g., converting landscape video into portrait), adjust camera angles and pacing in AI-generated scenes to create timelapses, drone-style clips, and more. These features are available in preview.
Furthermore, Imagen 3 image generator has seen significant performance improvements, allowing it to remove objects and reconstruct missing or damaged portions of images with better accuracy.
All media generated by Imagen, Veo, and Lyria models will be watermarked using Google’s SynthID technology. The company claims that its generative AI models have built-in safeguards to protect against the creation of harmful content.
While some companies in the industry train their models on copyrighted works without obtaining permission from rights holders, Google has not disclosed which specific data it uses to train its models and maintains a tradition of sticking with this approach.
Source: https://techcrunch.com/2025/04/09/google-brings-a-music-generating-ai-model-to-its-enterprise-cloud/
