←
AudioShake808 posts
AudioShake avatar

AudioShake

@AudioShakeAI

Making audio more usable for humans and machines.

◷ Joined October 2019▧ 386 Photos and videos
AudioShake avatar
AudioShake@AudioShakeAI·Sep 1

Multi-Speaker 2.0 is live–the world’s first low and high-resolution multi-speaker separator. One recording of several people in. A clean labeled track per person out, including the moments when two people talk at once. 8 kHz to 48 kHz, one model. Confidence scores every 20ms.

0:53
AudioShake avatar
AudioShake@AudioShakeAI·20h

Mirelo’s new editor assistant reworks a video’s sound when you ask it to. @AudioShakeAI is the separation layer underneath: it pulls out the music, isolates dialogue and splits speakers, then Mirelo builds the new sound. Congrats @MireloAI 👏

We cracked agentic sound design. Introducing Mirelo Sound Agent. Add sound effects, dub voices and swap music on your video, just by prompting (thread) Try it: mirelo.ai/audio-editor

0:55
AudioShake avatar
AudioShake@AudioShakeAI·Nov 16

Two people talk at once. Transcripts fall apart. Multi-Speaker 2.0 gives every speaker a clean track, overlap included: 4.1× fewer transcription errors than top open-source, 30% lower diarization error than pyannote. #AudioAI #SpeechRecognition

0:29
AudioShake avatar
AudioShake@AudioShakeAI·Nov 12

Today at @SVGEurope’s Sports Audio Summit: Mike Arthur on source separation for live sport. 12:20pm GMT, King's Place, London. Find him at our booth after.

Post attachment
AudioShake avatar
AudioShake@AudioShakeAI·Nov 5

A 1998 VHS tape of ODB freestyling ended up on SZA's SOS. Rodney Jerkins: "I took the VHS, turned it into audio, put it through AudioShake, and then just pulled the vocals out." audioshake.ai/case-studies/r…

0:55
AudioShake avatar
AudioShake@AudioShakeAI·Nov 4

Next Thursday at @SVGEurope's Sports Audio Summit in London: Mike Arthur on source separation for live sport. 12:20pm GMT, Kings Place. Come say hi at our booth.

Post attachment
AudioShake avatar
AudioShake@AudioShakeAI·Oct 29

A finished mix doesn’t have to be final. Sun 11/1, 1pm at #AESShow Nashville: Suzanne Kirkland on the Focusrite stage on how Judy Garland’s vocal was separated from the original mono of The Wizard of Oz for Sphere. Plus John Merchant on the Bee Gees 🎶

Post attachment
AudioShake avatar
AudioShake@AudioShakeAI·Oct 29

A finished mix doesn’t have to be final. Sun 11/1, 1pm at #AESShow Nashville: Suzanne Kirkland on the Focusrite stage on how Judy Garland’s vocal was separated from the original mono of The Wizard of Oz for Sphere.

Post attachment
AudioShake avatar
AudioShake@AudioShakeAI·Oct 27

Headed to #AESShow Nashville, Oct 30 – Nov 1. Bring a mix you’re stuck on to booth #521, and watch us separate it for you.

Post attachment
AudioShake avatar
AudioShake@AudioShakeAI·Oct 26

Expired music license? Detect it, identify it, remove it, replace it. Jessica Powell with @AWS at #IBC2026

0:50
AudioShake avatar
AudioShake@AudioShakeAI·Oct 22

Heading to Sacramento Tech Week? Swing by RELAY 001 tomorrow. Attendees get AudioShake credits to separate stems and work with source audio in new, creative ways. Fri 10/23 · 5–8 PM PT RSVP: luma.com/erxl3n30

Post attachment
AudioShake avatar
AudioShake@AudioShakeAI·Oct 22

What does a Lite-Brite have to do with AI? A finished mix is one picture made of many pegs. Source separation pulls every peg of one color (vocals, dialogue, effects) onto its own board. @themoko explains 👇 audioshake.ai/post/what-is-a…

3:58
AudioShake avatar
AudioShake@AudioShakeAI·Oct 21

Accessibility isn’t a someday project anymore, and that includes the archive. Captions, audio description, and dubbing all need dialogue and M&E on separate tracks. Most catalog titles only have the finished mix. We separate them, so the work can start. 👇

Post attachmentPost attachmentPost attachmentPost attachment