AudioShake

AudioShake is an AI-powered platform that specializes in audio separation technology. It enables users to deconstruct audio files into component stems such as vocals, instruments, and effects, which can be utilized for various creative and commercial applications in the music and media industries.

Key features:

- Stem Separation: Splits music into individual instrument stems for mixing and mastering purposes, particularly useful for live recordings or tracks without original stems.
- Immersive Audio Support: Facilitates the creation of immersive audio experiences like Dolby Atmos and Sony 360 mixes.
- Dialogue Extraction: Enhances dubbing workflows by isolating clean dialogue from noisy backgrounds, improving transcription accuracy.
- Interactive Audio: Enables the creation of interactive experiences where users can manipulate individual audio elements.
- Sync Licensing: Provides high-quality stems to streamline the process of obtaining clearance for music usage in media productions.
- Lyric Transcription and Alignment: Offers automated, high-accuracy lyric transcription and alignment for creating karaoke tracks and lyric videos.
- Audio Analysis: Supports tasks such as sonic branding and content recognition through advanced separation capabilities.
- A/V Editing: Facilitates advanced audio/video editing by allowing separation of audio components, useful for projects involving copyrighted music or enhancing dialogue clarity.

How it works:

1. Users upload an audio file to the AudioShake platform.
2. AI-driven algorithms identify and isolate different components of the audio file.
3. The platform separates the audio into distinct stems (e.g., drums, vocals, background noise).
4. Users can access and download the separated stems for further use.

Integrations:

Chordal, Dubverse, cielo24

Use of AI:

AudioShake employs deep learning techniques to achieve high-quality audio separation. The AI algorithms are designed to identify and isolate different components of an audio file with precision.

AI foundation model:

The platform is based on a deep learning model specifically designed for 'source separation' tasks. This model has been recognized for its superior performance in competitions such as Sony's Demixing Challenge.

Target users:

- Music producers
- Audio engineers
- Content creators
- Media professionals

How to access:

AudioShake is available as a web app and API. Professional users can access AudioShake Pro for commercial purposes, which includes a vast catalog of songs and custom stem requests. Independent artists can use AudioShake Indie, which offers free experimentation with audio separation on up to 10 tracks per month.