A new partnership enables the creation of avatars fully integrated into the device, with latency under 100ms paving the way for realistic and evolving AI characters for games, film, and interactive media.
Neuphonic and Rapport (a division of Speech Graphics) announced a partnership to deliver what they believe will be one of the first photorealistic, anthropomorphic digital systems operating entirely in real time on standard CPU hardware.
This collaboration combines Neuphonic ‘s ultra-low latency speech synthesis technology with Rapport , an interactive AI platform for role-playing via avatars, which leverages Speech Graphics’ production-proven facial and body animation technology used in video games such as Call of Duty , The Last of Us Part II , and Hogwarts Legacy . Together, the system generates a natural voice, perfectly synchronized lip movement, and emotionally responsive facial expression in under 100 milliseconds, allowing digital characters to speak, react, and express emotions instantly, without any perceptible delay.
This paves the way for a new generation of interactive content, from live digital characters and real-time virtual production to reactive experiences in games and advertising – helping to overcome the uncanny valley effect and setting a new benchmark for realism, speed and accessibility for digital humans.
“We are making real-time digital humans accessible,” said Sohaib Ahmad, CEO and co-founder of Neuphonic. “For the first time, it’s possible to instantly run a fully interactive and lifelike avatar on consumer hardware. This moves the technology from a solution requiring high-end infrastructure to one that can work anywhere, and will pave the way for entirely new use cases in gaming, media, and AI. It will bring us closer to AI emerging as a natural and seamless extension of human interaction.”
Also Read: The Dawn of Agentic Silicon: How Arm’s New AGI CPU is Rewriting the Rules of Computing
Speech Graphics Ltd, a market leader in audio animation, has spent more than a decade developing technologies to create realistic characters in video games and entertainment, including tools used in major AAA titles, whose applications now extend to real-time conversational experiences.
“Neuphonic emerged as a natural partner because our technologies are perfectly complementary,” said Gregor Hofer, CEO and co-founder of Rapport, a division of Speech Graphics Ltd. “They bring ultra-low-latency, on-device speech technology that runs efficiently on the CPU, while we bring real-time facial animation that has a proven track record in production. Together, this enables the creation of fully local digital characters that can speak, animate, and react instantly, without relying on the cloud, opening up new possibilities in terms of performance, cost, and large-scale deployment. It’s a complete, real-time conversational experience.”
This technology paves the way for immediate applications across all sectors: in video games, by enabling NPCs to speak and react dynamically; in film and virtual production, by offering live voice and facial performances; and in advertising and customer engagement, by powering interactive digital spokespeople. Beyond that, areas such as education and virtual assistants stand to benefit from more engaging and responsive digital characters at scale.
Because the system operates entirely on the device, user data never leaves the environment, reducing latency, costs, and infrastructure complexity while making real-time avatars deployable on billions of devices.
This partnership was supported by Innovate UK’s Creative Catalyst program, which recognizes the growing importance of real-time, AI-driven digital humans in the creative and technology sectors.
Source: Businesswire
