Innovative Machine Learning Technologies Offer Audio Extraction Solutions for Music and Speech

PARIS – ​Audionamix is bringing its most advanced audio separation technologies to the worldwide stage at ​VivaTechnology 2019 (Stand B19-007). Built using artificial intelligence processing, Audionamix technologies can pinpoint specific speech​ and melodic elements, including vocals, drums, bass and remaining music.​ These proprietary technologies, ​which have been used in award-winning, commercially available software products and are now available for licensing, provide the ability to address both common and complex audio issues and to create stems for live performance and samples for music production.

Audionamix’s latest breakthrough in music separation technology allows mono or stereo recordings to be split into four stems – vocals, drums, bass and the remaining music. This brand-new technology, offering efficient separation of tracks while retaining the quality of each individual segment, will be featured at VivaTechnology. XTRAX STEMS, Audionamix’s popular three-stem software technology, will be showcased at the booth as well.

Audionamix’s speech separation technology uses advanced machine learning to isolate spoken word from challenging background interference. The tech features integrated artificial intelligence based on deep neural networks that automatically detect and separate speech, regardless of the surrounding content, creating an elegant and easy solution to complex audio challenges. Powered by this technology, Audionamix’s IDC: Instant Dialogue Cleaner plug-in can completely remove complex varying background noises such as wind, birds, insects, cars, planes, music, room reverb and more with the simple turn of a knob.

“Our algorithms are trained to recognize vocals, dialogue, bass and drums, allowing them to accurately identify and separate out these elements from any audio file,” says Nicolas Cattaneo, head of R&D, Audionamix. “Both the music- and speech-based solutions provide the opportunity for users to adjust, fix, sample or completely reinvent a recording.”

A live demonstration of the latest advancements to these technologies will be available at the Audionamix VivaTech 2019 stand (B19-007) and representatives will be on-site to discuss licensing agreements. Commercially available versions of its music (XTRAX STEMS 2) and speech separation (IDC: Instant Dialogue Cleaner) technologies will also be available. For more information, visit ​audionamix.com​.