Skip to main content

Introduction

AudioShake is a cutting-edge audio research company specializing in high-quality sound separation and lyric transcription technology. Stems are essential components of modern content workflows, opening up audio to new uses, while lyrics play a critical role in the music industry—powering sync briefs, DSP recommendations, and fan engagement.

Our proprietary, patented technology—including state-of-the-art lyric transcription—has processed millions of minutes of audio to create industry-leading solutions for music, film, television, UGC, gaming, streaming, sports, and more. AudioShake makes it possible to access and isolate the different components of a piece of audio for more precise editing, mixing, and creative use. See how our customers and partners are using AudioShake. See how customers and partners have used our technology.

To begin building with AudioShake:

  1. Create an account and API key via the API Dashboard.
  2. Purchase credits and use the task route to request stems.

For questions or support, contact us at support@audioshake.ai.

Platforms​

API

Use the API dashboard to create an account, manage keys, and monitor usage, then start building with the /task route to generate stems, lyrics, and more.

SDK

Bring real-time stem separation to edge devices with a lightweight SDK that delivers isolated stems in real-time.

Widget

Quickly perform stem separation and transcription with AudioShake’s embedded javascript pop-up tool.

Services​

  • Instrument Stem Separation: Separate any track–even mono-track recordings–into stems and instrumentals, for use in sync licensing, immersive mixes, AR/VR, gaming, fitness, UGC, karaoke, and more.
  • Dialogue, Music, and Effects Separation: Isolate dialogue, music and effects tracks for a variety of use cases across film, TV, dubbing, and synthetic voice.
  • Lyric Transcription and Alignment: Transcribe and word-align audio files into readable text, using AudioShake’s leading vocal isolation models.
  • Multi-Speaker Separation: Isolate individual voices from overlapping speech in podcasts, film, and TV content. Create separate speaker streams for each speaker for use in dubbing, accessibility, voice AI, and more.