Unlock Real-Time Sound with AudioShake SDK

AudioShake is excited to announce our first-ever SDK, now available for iOS/MacOS, Android, Windows, and Linux. With just a few lines of code, developers can embed the same AI audio models used by leading music labels, broadcasters, and tech platforms—now optimized for real-time applications on edge devices.
Built with our best-in-class models, AudioShake’s SDK can separate vocals from a full mix, isolate speech from background noise, or pull out an individual instrument—all in real time. Whether you're building a music app, a transcription engine, or an educational tool, AudioShake’s SDK lets you isolate stems with unmatched quality and speed. Read more about the specifications of our SDK on our Developer Portal.
Built for Developers. Trusted by Industry Leaders.
With AudioShake’s SDK, sound becomes something users can shape, clean, remix, and control—instantly. AudioShake’s real-time SDK is already being used by streaming platforms, captioning services, and songwriting tools to make mixed audio more editable, accessible, and useful. Some of our earliest customers have helped shape the SDK, and are bringing real-time stem separation experiences to consumers and businesses:
Interactive Music Mixing
Leading music platforms are using Audioshake to empower their users with cutting-edge tools. For example, Tuned Global—a leader in white-label music streaming solutions—uses Audioshake’s SDK to enable clients to offer features like karaoke, remixing, and isolated stems, enhancing creativity and boosting user engagement within their platforms.
To see stem separation at work for consumers, check out apps like MashApp, which allow everyday music fans to create high-quality stems from popular songs and create their own mashups with ease. As one of the only apps to be licensed by major labels for music mixing, MashApp gives its users the freedom to open up a breadth of music to new remixes.
Similarly, vocal effects app Voloco uses AudioShake to separate vocals from instrumentals in existing commercial tracks. This lets users to sing over professional backing tracks to create polished performances and unique interpretations of the songs they love, all from a mobile device.
Speech & Live Broadcast Workflows
In the world of speech workflows, audio quality is everything. A bit of background musicor crowd noise can throw off even the best Voice AI and automatic speech recognition (ASR) models. That’s why AI Media, which provides live captioning solutions for any broadcast content, plus events like AWS re:Invent, integrated AudioShake’s SDK into their workflow. Using our SDK, AI-Media’s system can separate a sports commentator’s voice from the background crowd noise in real time. The isolated commentary is then fed into their transcription and translation pipeline (LEXI Voice) to produce accurate captions in real time.
Songwriting and Vocal Effects Tools
When a new idea hits, songwriters often turn to vocal recording apps to capture it fast. Clean vocals allow artists to fine-tune performances, collaborate, or experiment with effects—even from a phone. The AudioShake SDK is also fueling the next generation of creative tools for songwriters. Take VoCap, a mobile recording app from the team at AutoTune, which designed an app for songwriters to record and polish vocals right from their mobile device. By integrating AudioShake’s SDK, VoCap now lets users create clean vocal takes directly on their phone, no matter the recording environment. Users can also separate vocals from the melody on the same audio file for more customized mixing.
Music Education and Practice
On-demand stem separation can also turn a mobile app into a virtual rehearsal studio. Backtrackit, a popular practice tool for musicians, uses AudioShake’s SDK to make learning songs easier and more interactive. Its “Track Splitter” interface allows users to create karaoke-style tracks in order to isolate tricky parts of songs, loop them, slow them down, and really hear every nuance.
Get Started with AudioShake SDK
With AudioShake’s SDK, you can treat any song, recording, or stream as a dynamic, malleable experience – muting, enhancing, and reimagining elements on demand. Whether you're shipping a consumer music app or a professional tool for broadcasters, AudioShake’s SDK gives you the power to work with clean, editable audio in real time.
To our developer community, we invite you to join us in exploring what’s possible. The AudioShake SDK is available now for iOS/MacOS, Android, Windows, and Linux. Ready to build the future of sound? 👉 Visit developer.audioshake.ai to get started. We can’t wait to hear what you create.