LALAL.AI Launches Lynx Voice Cleanup Mode

LALAL.AI, providers of an artificial intelligence-powered audio processing platform, today launched Lynx, a neural network exclusively for speech denoising.

Lynx is trained to separate speech from background music, crowd noise, mechanical interference, environmental sounds, and the full range of acoustic artifacts present in content produced outside controlled studio conditions. The result is a clean voice track ready for further post-production without manual cleanup.

"In speech denoising, you're not separating things that were recorded together by design. You're trying to recover a voice from an environment that was never meant to be a recording studio," says Nik Pogorsky, LALAL.AI's product owner and co-founder, in a statement. "We spent a year building a model that treats that as the actual problem, not a side case of something else."

The model is six times smaller than Andromeda, LALAL.AI's flagship cloud music stem separation model. Lynx was trained on a manually curated dataset of audio tracks covering conditions from quiet interviews to noisy field recordings. 

Lynx is available now through LALAL.AI's Voice Cleaner and Voice & Noise stem via browser, mobile app, desktop cloud app, and the LALAL.AI API. Planned improvements to the Lynx architecture include enhanced separation of choral and group vocals and improved isolation of speech recorded at a distance from the microphone.