.png){kind=logo}
Audio technology company L-Acoustics is using artificial intelligence to tackle one of live music's oldest problems: keeping vocals clear when everything around the microphone is loud. Its Source Intelligence technology uses machine learning to separate a performer's voice from unwanted sounds, including background instruments, crowd noise and reflections from the venue. The system is being used in major live productions, including Céline Dion's Paris residency, while audio specialists working on Harry Styles' tour have also been involved in its development.
At a concert, microphones pick up more than the singer's voice. Drums, guitars, loudspeakers and the acoustics of a large venue can all interfere with the vocal mix. Sound engineers traditionally use filters and other processing techniques to reduce this interference, but those adjustments can affect the natural sound of a performance.
Source Intelligence takes a different approach. Its machine-learning system identifies the voice and removes unwanted audio in real time, helping engineers produce cleaner vocals without relying as heavily on conventional processing.
According to L-Acoustics, the technology can reduce non-vocal signals by up to 40 decibels. It runs on the company's L-ISA Processor II and is designed to work within its own audio system.
Also Read: How Spotify’s Founder Daniel Ek is Using AI to Transform Music Discovery
Céline Dion's return to live performance has brought renewed attention to the technology behind her shows. During her Paris residency, engineers have used a carefully arranged speaker system to deliver consistent sound across the arena.
The setup includes dozens of speakers positioned around the venue, helping audiences hear the performance clearly from different seating areas. Source Intelligence adds another layer by reducing unwanted noise around the lead vocals.
The aim is straightforward: preserve the detail and emotion in Dion's voice without letting the scale of the venue overwhelm it.
L-Acoustics has also identified Solotech, the production company working on Harry Styles' tour, as a beta partner for Source Intelligence. The collaboration reflects growing interest in using AI to solve practical audio challenges during large-scale live performances. For touring engineers, clearer vocals can mean fewer compromises when performers move around the stage or sing close to loudspeakers.
