Nuance Secures $50 Million to Develop Real-Time Audiovisual AI Models

Industry Pulse News Desk · 2026-09-14

Seattle startup Nuance has raised $50 million in Series A funding to build AI models capable of seamless, face-to-face conversational interactions.

Seattle-based artificial intelligence startup Nuance has raised $50 million in a Series A funding round led by Lightspeed Venture Partners. The round, which follows a $10 million seed funding effort last year, included participation from Accel, Nvidia's NVentures, South Park Commons, and Define Ventures.

Founded by three former Apple researchers, the startup is developing an integrated audiovisual-to-audiovisual AI model designed to make digital conversations more natural. Current voice chatbots typically combine separate speech-to-text, large language processing, and text-to-speech models, which often creates latency and leaves digital avatars unable to respond dynamically to human non-verbal cues.

Nuance's single-system approach processes simultaneous video and audio streams for both input and output. Because the model requires specialized training material, the company collects data directly through internal recordings and specialized partners rather than scraping public internet sources, capturing two-way conversations with cleanly separated audio and video tracks.

The company plans to use the fresh capital to expand its eight-person team, focusing on research and development alongside go-to-market strategy hiring. While Nuance remains pre-product, it intends to release a public research preview later this year, targeting business-to-business applications such as virtual sales representatives, automated customer support, and interactive interview tools.