Smallest.ai Raises $13M to Build Ultra-Fast Voice AI That Sounds Human

Founded in late 2024, Smallest.ai has successfully secured $13 million in a Series A funding round, led by Seligman Ventures, with additional contributions from Sierra Ventures and 3one4 Capital, raising its total financing to over $21 million. The startup is creating a compact voice model that aims to replicate human-like conversation by simultaneously listening, thinking, and speaking, striving to make AI indistinguishable from humans. Founder and CEO Sudarshan Kamath notes that their model allows for real-time interaction with minimal response delays, unlike traditional large language models (LLMs). When faced with unfamiliar topics, the model briefly places the customer on hold to consult a larger foundational model, mimicking human behavior. Smallest.ai emphasizes voice-specific details, accommodating various accents, supporting multiple languages, and functioning in noisy settings. Its clients include RingCentral and Truecaller, and it competes with ElevenLabs, Cartesia, and local firms like Sarvam. Kamath envisions a future where AI agents utilize both a small voice model for immediate engagement and an 'offline' LLM for intricate queries. He asserts, "We want our models to break the Turing test. You should speak to our model and not know it's AI or human. That's the sole focus of the company."