Meta Reportedly Acquires AI Audio Startup WaveForms Amid Ongoing Talent War

Meta is reportedly continuing its recruitment offensive to build up its new artificial intelligence (AI) division Superintelligence Labs. As per the report, the Menlo Park-based tech giant’s latest spoils of the war is WaveForms, an AI startup working with audio-focused models. While the terms of the acquisition were not disclosed, the co-founders of the company were reportedly hired by Meta. This comes just a month after the company reportedly acquired Play AI, another startup working in the AI audio space.

WaveForms Was Founded in December 2024

According to The Information, Meta has completed the acquisition process to purchase WaveForms, an audio large language model (LLM) research and product startup. On LinkedIn, the company says it is working to solve the Speech Turing Test and creating “Emotional General Intelligence.”

Notably, the Speech Turing Test is a variation of Alan Turing’s original Turing Test that evaluates whether a listener can tell if a voice or speech is generated by a human or by AI. If the listener cannot confidently pick an option, the test is considered to be passed. It is considered one of the important benchmarks of voice-based AI.

The San Francisco-based AI startup was founded by ex-Meta and OpenAI researcher Alexis Conneau, and ex-Google advertising strategist Coralie Lemaitre in December 2024. Despite its short run, the company reportedly raised $40 million (roughly Rs. 350 crore) in funding earlier this year, in a round led by venture capital firm Andreessen Horowitz.

As per the report, both Conneau and Lemaitre will join Meta’s Superintelligence Labs as part of the acquisition deal. The tech giant might also make use of the tech stack of WaveForms to develop its voice-focused AI products. Meta already owns Play AI, another voice AI startup it acquired last month.

The company’s focus on targeting voice AI startups highlights its intention to fill a visible gap in Meta AI’s capabilities. While the company is reportedly working on adding real-time two-way voice conversations with the chatbot, it remains significantly behind compared to competitors like OpenAI and Google, which have released natural sounding voice features within ChatGPT and Gemini, respectively.