AudioShake ran Multi-Speaker 2.0 against 15 public test sets including meetings, phone calls, broadcast, and web video to produce lower labeling errors than the leading open-source diarization system.
Published
Signal category
Research & Knowledge
Quote
“We ran Multi-Speaker 2.0 against 15 public test sets — meetings, phone calls, broadcast, web video — and it produced lower labeling error than the leading open-source diarization system on every one of them.”
— AudioShake team
Company
AudioShake
AudioShake helps make sound editable, accessible, and useful by separating sound into its different components.
- Industry
- Software Development
- Location
- San Francisco, US
- Company size
- 41 employees
Award-winning sound separation. Make audio interactive, editable, and customizable. AudioShake is a powerful AI that can take a recording from any point in time and deconstruct it into its component parts--from music instrument stems through to separate dialogue, music, and effect tracks. Creating stems opens up creative opportunities to make music and other audio more interactive, accessible, and profitable. - Winner of TIME Best Inventions - Winner of the Sony Demixing Challenge - Winner of the Music Tectonics and Wallifornia Startup Pitch Contests - Named by Business Insider as one of 14 Innovative Music Tech Startups
Founded 2020