The top talent behind yet another promising AI startup was swallowed up by the incumbent company. As part of the new licensing deal, Google DeepMind will bring in the CEO and several top engineers from voice AI startup Hume AI, Wired reports.
What remains of Hume AI will continue to provide its technology to other AI companies. Financial details of the transaction were not disclosed.
TechCrunch reached out to Google and Hume AI to confirm this news.
According to Wired, CEO Alan Cowen and about seven other engineers will work with DeepMind to improve Gemini’s voice capabilities.
Hume’s acquisition is the latest example of a major AI company taking talent out of the market and avoiding regulatory scrutiny by buying teams at startups rather than buying companies outright. Last year, Google acquired the CEO and other top researchers of viral AI coding startup Windsurf, and OpenAI has acquired several startup teams in recent months, including Covogo and Roi. The Federal Trade Commission recently announced it would take a closer look at these transactions.
The deal also shows that voice is becoming the next frontier for AI.
The secret to Hume AI is the model’s ability to understand your emotions and mood based on your voice. In 2024, the startup launched Empathetic Voice Interface, a conversational AI with emotional intelligence. Hume AI has raised nearly $80 million to date, according to PitchBook, and is expected to bring in $100 million in revenue this year, according to Wired.
tech crunch event
san francisco
|
October 13-15, 2026
But Hume AI isn’t the only company working on voice-focused models. Google has been steadily improving its Gemini Live feature, which allows users to converse with chatbots. According to the Gemini API release notes, Google released a new native audio model for the Live API last month. This increases the model’s ability to “handle complex workflows”.
Other companies in the industry are investing heavily in voice capabilities as well. OpenAI is reportedly preparing to overhaul its audio model in preparation for the release this year of audio-first personal devices made with Jony Ive’s IO. Recent leaks suggest that the device could be a type of earphone.
Last year, Meta accelerated its AI audio efforts by acquiring startup Play AI. The Facebook maker’s Ray-Ban smart glasses increasingly rely on voice and audio capabilities for tasks such as making it easier to hear conversations in noisy rooms and allowing hands-free control of calls, texts, music, and photos.
“Voice is the only input mode that wearables accept,” investor Vanessa Larco told TechCrunch. “This acquisition will further accelerate the need for voice apps.”
Demand for voice capabilities continues to increase. Earlier this month, AI voice generation startup Eleven Labs announced annual recurring revenue of more than $330 million.
