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

Smallest.ai, a startup founded in late 2024, has raised $13 million in a Series A round to advance its mission of making AI voice agents indistinguishable from human conversation. The funding round was led by Seligman Ventures, with participation from Sierra Ventures and 3one4 Capital, bringing the company’s total funding to over $21 million. The startup focuses on developing small, specialized voice models that enable real-time, natural conversations for enterprise customer support.

Why Smallest.ai’s approach differs from large language models

Traditional large language models (LLMs) process prompts in a batch, which introduces latency that feels unnatural in voice interactions. Smallest.ai’s model is designed to listen, think, and speak simultaneously, mimicking human conversational patterns. Sudarshan Kamath, founder and CEO, explained that while speaking, humans are already processing and may interrupt, a behavior the startup’s model replicates. This real-time intelligence layer reduces response lag to near zero, making conversations feel more natural.

Handling complex queries with hybrid model architecture

When the voice model encounters a subject outside its specialized knowledge, it hands off the query to a larger foundational model, briefly placing the customer on hold to ‘research’ the issue—similar to how a human agent might pause to look up information. Kamath believes all AI agents will soon rely on two models: a small voice model for real-time interaction and an offline LLM for complex problem-solving. This hybrid approach allows Smallest.ai to focus on voice-specific nuances like accents, multilingual support, and noisy environments.

Market position and competition

Smallest.ai competes with voice AI leaders like ElevenLabs and Cartesia, as well as regional players like Sarvam. Unlike competitors that apply voice AI to use cases like audio dubbing and podcasting, Smallest.ai focuses strictly on real-time conversational voice agents for enterprise customers. The startup’s existing clients include RingCentral and Truecaller, and it sees potential customers in the broader customer support space, including newer AI companies like Sierra and Decagon.

Why this matters for customer support

The push for more natural voice AI addresses a common pain point: customers can often tell they’re speaking to a machine. By reducing latency and mimicking human conversational cues, Smallest.ai aims to improve customer experience and operational efficiency. For businesses, this could mean higher satisfaction and lower costs, but it also raises questions about the future of human jobs in customer support. As Kamath put it, ‘We want our models to break the Turing test.’

Conclusion

Smallest.ai’s Series A funding underscores growing investor confidence in specialized voice AI solutions. With a focus on real-time interaction and a hybrid model approach, the startup is positioning itself as a key player in the evolution of conversational AI. As the technology matures, it could redefine how businesses handle customer interactions, making them more natural and efficient.

FAQs

Q1: What is Smallest.ai’s core technology?
Smallest.ai develops small, specialized voice models that process speech in real time, allowing AI agents to listen, think, and speak simultaneously, reducing latency to near zero.

Q2: How does Smallest.ai handle complex queries?
When the voice model encounters a query outside its knowledge, it hands off to a larger foundational LLM, briefly placing the customer on hold to ‘research’ the issue, similar to human behavior.

Q3: Who are Smallest.ai’s main competitors?
Competitors include ElevenLabs, Cartesia, and regional players like Sarvam, but Smallest.ai differentiates by focusing exclusively on real-time conversational voice agents for enterprise customer support.