Salsa Sound are a UK-based startup producing AI-driven software for assistive mixing and audio quality control of live sports and entertainment – dramatically improving audio quality for viewers and reducing costs for broadcasters.

Our MIXaiR product improves live production efficiency, capturing all key sounds to deliver fully immersive content – engaging audiences with better storytelling. Complementing this, our quality control probes can be inserted in the signal path for automated monitoring and intelligent algorithmic analysis. Our QC probes cover speech intelligibility, audio fidelity, multi-channel phase/level analysis, profanity filtering, language detection and more.

MIXaiR runs on Windows, Mac and Linux with an API backend and browser-based UI which can also be seamlessly virtualised in the cloud. The intuitive drag-and-drop approach enables multiple, simultaneous mixes to be configured in a fraction of the time compared with traditional approaches.

MIXaiR supports routing multiple simultaneous inputs/outputs across any format (MADI, ASIO, Core Audio, Dante, SDI, SRT, NDI, 2110, etc.) and automates the creation of sub-mixes (field-of-play, audience, speech, etc.), optionally processed with VST plugins. Sub-mixes can be added to any number of output mixes any format (from stereo, 5.1, 5.1.4, 22.2, ambisonics etc. to MPEG-H and Dolby ATMOS)—producing multiple high-quality (multi-language) mixes simultaneously, all automatically adjusted to the correct loudness standard.