Artificial intelligence (AI) is getting a higher profile across two of Google’s premier product lines.
Gemini, the family of multimodal large language models (LLMs) developed by Google DeepMind, is replacing Google Assistant on Android phones later this year. Meanwhile, Chirp 3 — the company’s speech-to-text and HD text-to-speech models — are being added to the Vertex AI development platform this month.
Both moves offer another vivid example of how tech’s biggest players are deeply incorporating AI into existing products amid a clamor for the technology from enterprises and consumers.
Google said it will upgrade users from Google Assistant to Gemini “over the coming months.” Assistant will no longer be accessible on most mobile devices or available from app stores later this year, it added.
“We’ll be upgrading tablets, cars and devices that connect to your phone, such as headphones and watches, to Gemini,” the company said in a blog post last week. “We’re also bringing a new experience, powered by Gemini, to home devices like speakers, displays and TVs.”
As it makes the transition, Google is adding new features to Gemini on Android devices, such as the ability to play music and an option to take actions directly from a user’s lock screen. The move to Gemini isn’t surprising and indeed seemed inevitable: Google launched its Pixel 9 smartphone line with Gemini as the default virtual assistant in August, noting its more advanced capabilities in getting help and information via tools like Gemini Live and Deep Research.
The addition of Chirp 3 to Vertex is equally logical. A key focus of generative AI has been text-based interfaces used to create text, images and — increasingly — voice.
A handful of major startups have emerged as threats to Google in the space. Last week, Sesame announced the launch of its model for developers to build customized apps and services. And in late January, ElevenLabs said it raised a round of $180 million, pumping its valuation to $3.3 billion.
Last week, Google said Chirp 3 would introduce eight new voices for 31 languages. At a news conference at Google DeepMind’s offices in London, the company said use cases would include building voice assistants, creating audiobooks and developing support agents and voice-overs for videos.
Chirp 3 will come with restrictions to minimize misuse of the technology. “We’re just working through some of these things with our safety team,” Google Cloud CEO Thomas Kurian said.
