On Monday, Google announced that its Chirp 3 artificial intelligence (AI) model will be integrated into the Vertex AI platform. Previously available only in private preview, the model is now set to be accessible to all users of the AI platform. Chirp 3 specializes in audio generation, featuring a variety of custom voices that emulate human-like inflections and intonations. Furthermore, the Mountain View tech company recently enhanced the model, adding eight new voices in a total of 31 languages.
Chirp 3 AI Model Is Now Available in Vertex AI
In a press release, Google Cloud shared details about the availability of the Chirp 3 AI model via its Vertex AI platform during the “Gemini for the United Kingdom” event held at Google DeepMind’s headquarters in London. The occasion also featured a number of other AI-centric announcements from the company’s Cloud division.
Chirp 3 is expected to be accessible on Vertex AI starting next week, joining other advanced models available on the platform, including Gemini, Imagen, and Veo. Upon launch, the Chirp 3 HD Voices feature will support 31 languages and provide 248 unique voices with eight different speaker options. Google asserts that Chirp 3 delivers custom speech generation with human-like intonation and emotional nuances.
The technology is being promoted for a range of applications, including voice annotation, real-time transcription of meetings, audiobooks, podcast narration, and gathering sentiment feedback from customer service interactions. Additionally, it can be utilized in creating voice assistants and AI agents equipped with voice support.
According to a report from TechCrunch, certain usage restrictions will be imposed on the Chirp 3 AI model to minimize the potential for misuse. Thomas Kurian, Google Cloud’s CEO, indicated during the event that the company is focusing on implementing safety measures before broadening access to the model.
Vertex AI serves as Google’s developer-centric cloud-based platform for building, deploying, and scaling AI models. It provides a seamless experience for deploying and testing new models, as well as for developing applications based on these models. While the tech giant provides one-time free credits for initial exploration of the platform, ongoing usage will necessitate a subscription.
