AI voice startup ElevenLabs and Alphabet’s (GOOG) (GOOGL) Google Cloud announced a multi-year extension of their collaboration to make AI voice tools more accessible to businesses globally.
ElevenLabs will use Google Cloud’s G4 virtual machines, or VMs, powered by Nvidia’s (NVDA) RTX PRO 6000 Blackwell Graphics Processing Units, or GPUs, to train and serve its voice models.
The companies said that ElevenLabs is also integrating Google’s Gemini AI models directly into its Agents Platform to unlock reasoning and multi-step planning for its voice assistants.
The startup is also incorporating Google’s Veo model into its Creative Platform, enabling teams to produce multimedia content — including video and audio — in less time, according to the companies.
The new agreement provides access to a significantly larger cluster of Nvidia’s Blackwell GPUs, enabling ElevenLabs to reliably support even larger deployments for enterprise customers and ensure its research teams have access to highly optimized AI compute.
Under the collaboration, ElevenLabs has also launched its solutions on Google Cloud Marketplace, allowing customers to scale conversational agents for customer support, internal training, and inbound sales.
“Now with G4 VMs powered by NVIDIA Blackwell, we’re pushing our multimodal models even further—faster inference, better reliability, instant replies across languages,” said ElevenLabs’s Co-Founder Mati Staniszewski.
Earlier this month, ElevenLabs raised $500M in a Series D funding round led by Sequoia Capital, valuing the company at $11B.