AudioShake Refinery
AudioShake Refinery separates overlapping voices into labeled audio tracks.
Find alternatives to AudioShake Refinery
What it does
- Audio editing
Features 6
- Corpus DiagnosisAssesses what audio is usable, fixable or junk before compute is spent on training.
- Multi-Speaker SeparationSplits multi-speaker conversation, including overlapping speech, into individual speaker tracks.
- Instrument StemsIsolates 10+ individual instruments such as vocals, bass, drums and guitar from mixed tracks.
- Music RemovalRemoves all background music, including sung vocals, leaving a pure speaker signal.
- Confidence ScoresEvery multi-speaker output ships with scores showing how cleanly a source is isolated.
- On-Prem DeploymentThe Refinery can be deployed inside the customer's own environment so audio never leaves.
Sources
- Pricing
-
Official documentation · 16 Oct 2026
"Contact us to learn more."
- Description
-
Official documentation · 16 Oct 2026
"can recover overlapping voices as separate, labeled audio tracks."
- What it does
-
Official documentation · 16 Oct 2026
"overlapping voices as separate, labeled audio tracks"
- Sold within
-
Official documentation · 16 Oct 2026
"Bring us the audio"
- Name
-
Official documentation · 16 Oct 2026
"AudioShake's Refinery works from finished, mixed recordings"
- URL
-
Official documentation · 16 Oct 2026
"AudioShake's Refinery works from finished, mixed recordings"
- Deployment
-
Official documentation · 16 Oct 2026
"Run on our API, or deploy inside your own environment"
- Interfaces
-
Official documentation · 18 Oct 2026
Page: 'Structured, training-ready data delivered over API or inside your own environment'.
- Intended for
-
Official documentation · 18 Oct 2026
Page: 'AUDIO DATA FOR AI TEAMS'.
Also from AudioShake
-
AudioShake Studio Application
Web app that separates a recording into vocal and instrument stems, splits dialogue, music and effects, gives each speaker a labelled track of its own, recovers speech from noisy or reverberant audio and transcribes and aligns lyrics word by word.
-
AudioShake API API service
REST API and SDK for AI source separation — vocals and instrumental, dialogue versus music and effects, and per-speaker tracks — billed in credits.
-
AudioShake Copyright Compliance System API service
AudioShake Copyright Compliance System detects, identifies, and removes music and generates cue sheet data.
-
AudioShake Live Application
AudioShake Live provides a web platform for separating dialogue, music, and effects from uploaded recordings.
Open AudioShake in the directory
Something wrong here? Send a correction — quote this product id: audioshake-refinery.