Nomono opens its audio engine to developers and AI agents

Nomono launched the Nomono Platform, an alpha API and MCP server that turns lavalier and ambisonics recordings into spatial, broadcast-ready audio scenes. The move opens the engine behind Nomono Studio to developers, producers and AI agents as the company targets immersive production and voice AI workflows.

Why it matters: - Nomono is trying to automate one of the hardest parts of immersive production: turning multi-person field recordings into clean, spatial audio that is ready for broadcast. - The platform also targets voice AI, where real-world speech data can improve how systems understand who is speaking, when they speak and where they are in a room. - The product could cut the manual post-production work that often takes sound engineers days per shoot.

What happened: - Nomono launched the Nomono Platform on November 5, 2026, in Warsaw, Poland. - The platform is the first product from Nomono Labs, a new effort to open Nomono’s audio technology to developers, partners and AI agents. - The company planned a live demonstration at Immersive Warsaw in a Dolby Atmos cinema. - An alpha program is already underway with selected immersive producers, educators and software makers.

The details: - Developers access the system through an API and an MCP server. - Producers upload lavalier tracks and an ambisonics recording from a shoot. - The platform separates each speaker, enhances the audio and places each voice where that person stood in the room. - Outputs include binaural mixes and ADM masters in Dolby Atmos, EBU and IAMF profiles. - The system isolates and cleans each voice to broadcast loudness standards. - It measures each speaker’s position from the ambisonics recording and tracks movement over time. - The platform reports who speaks and when. - Outputs also include processed and unprocessed files per microphone, stereo mixes, binaural mixes and spatial audio master files. - The files open directly in DaVinci Resolve and Pro Tools. - The platform works with third-party gear. - At launch, it accepts any lavalier kit plus a Sennheiser AMBEO or Zoom H3-VR ambisonics microphone. - Nomono says no other service automatically tracks speakers in a sound field. - Nomono says it holds granted patents on both known methods for doing so. - Speaker positions are delivered as standard ADM metadata. - That metadata can drive sound design in tools like Sound Particles. - The MCP server gives the same capabilities to AI agents including Claude, ChatGPT and Cursor. - Nomono said anyone can ask an agent to clean up a full shoot.

Between the lines: - Nomono is packaging the same processing chain used in its own Studio product into a broader platform play. - The company is betting that real-world audio structure will matter as much for AI as for post-production. - Its argument is that models trained on real speech in real environments should outperform systems built on synthetic approximations. - That approach could matter in applications such as healthcare, robotics and the smart home.

What’s next: - The alpha program is free during testing, with limited access for selected partners. - Nomono said pricing will be usage-based and set with alpha partners before commercial launch. - The company plans to extend the platform to post-production tools, archives and voice-AI products. - More information is available at the alpha program and the API documentation.

The bottom line: - Nomono is turning a niche immersive-audio workflow into a platform strategy, with an eye toward both faster production and better machine understanding of speech in the real world.