UK-based AI-based audio mixing specialist Salsa Sound has launched MIXaiR 2.0, the latest iteration of its automated platform for live audio mixing – which sees it transitioning to a pure software, cloud-ready service.
Aimed at live sports applications and built on proprietary AI technology, MIXaiR 2.0 allows sound engineers to automatically create an optimum immersive mix from standard microphone set-ups.
Regardless of whether viewers access content on a big screen or via social media, MIXaiR 2.0 offers multiple automated spatial audio mixes for headphones and loudspeaker set-ups. MIXaiR 2.0 requires no additional tracking or manual operation, instead taking audio feeds from existing broadcast microphones at a stadium and relying on its AI algorithms to detect, mix in and enhance the on-pitch sounds in real time.
MIXaiR 2.0 is designed to speed up audio workflows, automatically rendering to multiple formats and mixing multiple language versions or crowd ‘flavours’ (home/away), with each mix automatically made compliant to the requisite loudness standards and parameters required for social media platforms, linear broadcast, VOD or OTT.
‘We designed MIXaiR 2.0 to remove time-intensive, manual processes that sound engineers are faced with; instead, we give them the freedom to craft a mix, rather than chasing it,’ says co-founder and CEO, Rob Oldfield. ‘We want amazing sound to be accessible to as many sports fans as possible for both high and lower budget games. By removing the need for additional infrastructure, we’ve ensured that MIXaiR 2.0 can be used by clubs of all sizes to give their fans the best experience that captures all the excitement of the on-field action and makes them feel like they are actually in the stadium with the fans.’
With this version of MIXaiR, Salsa Sound has transitioned its technology from a 1U-high hardware platform to a pure software solution, supporting all major operating systems (MacOS, Linux, Windows) as well as VMs and major cloud platforms. This highly scalable solution is cost-efficient and flexible and can be deployed onsite, as part of a remote production workflow, over IP or in the cloud.
Through its AI-based approach, Salsa Sound aims to give a more robust and intelligent way to create automated mixes than conventional approaches. MIXaiR 2.0 has been ‘trained’ with hundreds of hours of content from English Premier League and Championship games to learn what sounds make up a great mix. The system automatically recognises the significant audio moments in the game and can automatically reduce profanity from pitch-side mics getting into the broadcast.
As well as for football, Salsa Sound has been trialling MIXaiR 2.0 with live boxing and is working to train the system with content from American Football and other sports.
