From our roots in a university research project, Salsa Sound is now ready to make waves across the pond with our patented AI technology for automating the creation of live audio mixes for sports.
MIXaiR™, now in its second iteration as an all-software, cloud-ready solution, was designed to automate current mixing processes and free up sound engineers so that they can craft a mix instead of chasing it – we are giving sound engineers the tools to create better content by allowing them to show off their skills and making it easier to create multiple simultaneous mixes. We want to give broadcasters and sports organisations the ability to create more content of a higher quality, without incurring additional costs.
MIXaiR™ was “trained” using countless hours of content from English Premier League and Championship games, and uses a proprietary process to intelligently deliver perceptually tuned power-matched crossfades, for seamless transitions when fading from one microphone to the other.
The engine relies on authentic audio from actual live sports that can be used to train the AI algorithms. The next step then is to feed individual audio stems, rather than fully mixed soundtracks, to enable the AI to analyse discrete sonic elements in American Football content. We’re in talks with some US football operators to provide the training content, and we’re excited about the real benefits they could see from automated/enhanced mixing.
Interested in what you read? Get in touch to find out more about Salsa Sound.