Coval Raises $28M Series A for Voice AI Testing

Coval, a new venture focused on building robust testing and evaluation infrastructure for voice AI agents, has successfully closed a $28 million Series A funding round. The investment was led by Norwest Venture Partners, signaling strong confidence in the company's mission to address critical gaps in the rapidly expanding voice AI market. This funding injection is set to accelerate Coval's development of simulation, observability, and human-in-the-loop systems designed to ensure the reliability and scalability of voice-powered applications.

The company was founded by Brooke Hopkins, who previously played a pivotal role at Waymo. At the autonomous vehicle pioneer, Hopkins led the evaluation infrastructure team, instrumental in proving the viability of self-driving technology through extensive simulations. Her experience highlighted the challenges of validating complex, non-deterministic systems that operate in unpredictable real-world environments. Hopkins identified a striking parallel between the early days of autonomous vehicles and the current state of voice AI, where enterprise adoption is outpacing the availability of reliable testing tools.

The voice AI sector is experiencing explosive growth, with the global market for voice recognition technologies estimated to have surpassed $22 billion. Enterprises are increasingly deploying voice agents, particularly in contact centers, where conversational AI is projected to save businesses $80 billion in 2026 alone. However, this rapid deployment presents significant operational hurdles. Unlike traditional software, voice agents are inherently non-deterministic, meaning they don't follow rigid scripts and can produce varied outputs even with minor changes to underlying models or user inputs. This complexity renders conventional testing methodologies, such as unit tests and manual QA, largely ineffective.

Coval aims to bridge this critical infrastructure gap by providing a comprehensive platform that mirrors the rigorous testing frameworks developed for autonomous systems. The platform's core components include advanced voice agent simulation capabilities, enabling companies to test agents against millions of synthetic scenarios before deployment. This allows for the discovery of edge cases and ensures regression testing, functioning as a vital release gate. The company's approach is designed to provide enterprises with the confidence needed to scale their voice AI initiatives without risking customer experience degradation.

Beyond pre-deployment testing, Coval offers real-time observability into live voice agent interactions. This feature allows businesses to monitor performance, detect anomalies, and track user responses across their entire call volume, catching issues that simulation might miss. The platform also incorporates a human-in-the-loop labeling system, where human reviewers validate complex scenarios. These validated cases then feed back into the simulation engine, creating a continuous improvement cycle that enhances the accuracy of automated evaluations and strengthens the overall robustness of the voice AI agents.

The strategic investment from Norwest Venture Partners underscores the market's recognition of the need for specialized infrastructure in the AI domain. As voice AI becomes a foundational element for customer engagement and operational efficiency, the demand for reliable evaluation and monitoring tools will only intensify. Coval's focus on building this essential operational layer positions it to become a key enabler for enterprises navigating the complexities of deploying and scaling advanced voice AI solutions.