AudioShake
Separate audio into production-ready stems and derive speech/lyric metadata through AI audio processing.
- Music label
- Content creator
- Developer
- music stems
- dialogue
- lyrics
- soundtracks
- Web
- Audio production · editorial fit
- Captions & accessibility · editorial fit
- Audio & podcasting · editorial fit
Overview
AudioShake separates finished audio into production-ready components: music stems, individual speakers, dialogue/music/effects beds, recovered speech, and lyric text. Studio presents those models in a browser workspace, while API, SDK, and private deployment options support production systems.
The Free Studio tier starts with 200 credits and MP3 export for files up to ten minutes. Paid Starter and Pro tiers replenish credits monthly, unlock every Studio tool and richer export formats, and accept files up to two hours; enterprise deployments add security, teams, and service terms.
Supported use cases
Create remix stems
Extract vocals and instruments from an existing master.
Prepare localization audio
Separate dialogue from a reusable music-and-effects bed.
Process archives at scale
Use programmatic or private deployment for a large audio collection.
Limitations
Credits and exports vary
Free use is a one-time starting balance with MP3 and file-length limits; larger capacity and lossless exports require paid access.
Key capabilities
Music and post-production separation
Splits instruments or separates dialogue, music, and effects from a mixed recording.
Speech and lyric recovery
Isolates speakers, recovers difficult dialogue, and extracts editable lyric text.
Developer delivery
Offers REST, batch, streaming, on-device, and private-deployment routes outside the Studio UI.
