San Francisco (September 22, 2026)—AudioShake has updated its software Multi-Speaker with a new V2.0 edition, bringing more features to its dialogue separation tool that can a single mixed conversation and splits every voice onto its own clean audio stem.
Key among AudioShake’s separation technology features is its ability to tackle Crosstalk—the moments when two guests finish each other’s sentences, someone laughs over a punchline, a lav mic catches the host sitting next to them and so on. Multi-Speaker 2.0 is designed to isolate the voices without destroying the natural delivery.
• Deep Dive into Eclipsa Audio at Mix Sound For Film & TV!
Using AudioShake’s Studio platform or API, editors can upload a mixed file and export a variety of options. Users can get discrete speaker tracks matched directly to the source sample rate; an isolated ambience stem containing the room tone, traffic and background noise, allowing editors to duck, sweeten or rebuild the soundstage and so on.
The software also provides targeted confidence markers; instead of listening through an hour of dialogue to audit where the separation held, the tool flags low-confidence moments down to a 20-millisecond frame. Editors can jump straight to the few seconds that need an ear check.
• Discover more pro audio news—get a free Mix SmartBrief subscription!
The new 2.0 edition reportedly provides sharper isolation in noisy environments with 32 percent less bleed versus version 1.0, removing more room noise and voice bleed. It is also said to provide full-band coverage (8 kHz to 48 kHz), and operates as a purely acoustic model rather than a predictive language model, preserving accented delivery, multilingual conversations, and overlapping vocalizations without hallucinated edits.
Multi-Speaker 2.0 is available in AudioShake Studio.
