Google Releases New Features For Its AI Audio & Video Generation Tools

Summary: Google has introduced a handful of new features for its in-house artificial intelligence platforms, including Imagen 3, Veo 2, Chirp, and Lyria. Google’s Veo AI video generator has received an interesting update that allows you to give it camera directions from different angles. These new features are the search engine giant’s latest push to corner the enterprise generative AI tools market.

Google has introduced a handful of updates for its first-party artificial intelligence-powered video and image generation tools at the Cloud Next 2025 event. There are new features for Google’s in-house AI platforms, including Imagen 3, Veo 2, Chirp, and Lyria. You can access these AI models through Google’s AI toolkit, Vertex AI Media Studio. It is a suite of generative tools that allow people to create polished videos from just a text prompt.

Google’s Imagen 3 and Veo 2 AI tools are getting interesting new features like camera directions

Talking about the Imagen 3 first, Google’s text-to-image generator now offers improved image generation capability. Furthermore, Google has updated it to provide enhanced inpainting functionality for reconstructing missing or damaged portions of a photo. Also, the new features are making it easy to remove unwanted objects from the photos.

As for the Veo 2 AI video-creation model, Google has enhanced it with a bunch of new editing and visual effects customization options. Google’s updated Veo 2 AI tool now allows you to give it specific camera directions in AI-generated scenes. You can also set the pacing to create time lapses, drone-style clips, and more videos. Also, it can interpolate between specified beginning and end frames.

Besides, the Veo 2 AI video generator is now able to remove background photos, logos, and objects from existing videos. Furthermore, it can also extend the frame of video footage. These new Veo 2 features are available in preview for enterprise users at the moment. With these features, Google is trying to make it easier for people to create cinematic-looking videos using Veo 2.

New updates for Lyria and Chirp 3 AI audio generators

Google has also made available Lyria, a text-to-music AI model, in private preview with an allowlist in the Vertex AI Media Studio toolkit. The tool allows users to generate complete, production-ready music assets with a text prompt. It simplifies creating a music track to serve as the background score for an AI-generated video from Veo 2.

Furthermore, Google has rolled out an “Instant Custom Voice” functionality for its synthetic speech model, Chirp 3. Google says that this feature can now generate “realistic custom voices from 10 seconds of audio input.” The tool can “accurately separate and identify individual speakers in multi-speaker recordings” for better transcription.

It’s important to mention that Google watermarks all media generated by the aforementioned AI tools using its SynthID technology. The search titan said all its generative artificial intelligence models have “built-in safeguards” to protect against the generation of harmful content.