# Unsupervised Source Separation By Steering Pretrained Music Models

arXiv: https://arxiv.org/abs/2110.13071 · HTML: https://mlsanity.com/paper/2110.13071/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 28 references that are in the map.
- First posted on arXiv 25 Oct 2021.

## Builds on (most cited first, 16 of 16)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Generating Diverse High-Fidelity Images with VQ-VAE-2](https://mlsanity.com/paper/1906.00446/) (2019)
- [Jukebox: A Generative Model for Music](https://mlsanity.com/paper/2005.00341/) (2020)
- [Music Source Separation in the Waveform Domain](https://mlsanity.com/paper/1911.13254/) (2019)
- [Cutting Music Source Separation Some Slakh: A Dataset to Study the Impact of Training Data Quality and Quantity](https://mlsanity.com/paper/1909.08494/) (2019)
- [Codified audio language modeling learns useful representations for music information retrieval](https://mlsanity.com/paper/2107.05677/) (2021)
- [The 2018 Signal Separation Evaluation Campaign](https://mlsanity.com/paper/1804.06267/) (2018)
- [Evaluation of CNN-based Automatic Music Tagging Models](https://mlsanity.com/paper/2006.00751/) (2020)
- [Automatic tagging using deep convolutional neural networks](https://mlsanity.com/paper/1606.00298/) (2016)
- [Simultaneous Separation and Transcription of Mixtures with Multiple Polyphonic and Percussive Instruments](https://mlsanity.com/paper/1910.12621/) (2019)
- [Finding Strength in Weakness: Learning to Separate Sounds with Weak Supervision](https://mlsanity.com/paper/1911.02182/) (2019)
- [Transcription Is All You Need: Learning to Separate Musical Mixtures with Score as Supervision](https://mlsanity.com/paper/2010.11904/) (2020)

## Built on by (most cited first, 1 of 1)

- [Zero-Shot Duet Singing Voices Separation with Diffusion Models](https://mlsanity.com/paper/2311.07345/) (2023)

## Abstract (arXiv)

We showcase an unsupervised method that repurposes deep models trained for music generation and music tagging for audio source separation, without any retraining. An audio generation model is conditioned on an input mixture, producing a latent encoding of the audio used to generate audio. This generated audio is fed to a pretrained music tagger that creates source labels. The cross-entropy loss between the tag distribution for the generated audio and a predefined distribution for an isolated source is used to guide gradient ascent in the (unchanging) latent space of the generative model. This system does not update the weights of the generative model or the tagger, and only relies on moving through the generative model's latent space to produce separated sources. We use OpenAI's Jukebox as the pretrained generative model, and we couple it with four kinds of pretrained music taggers (two architectures and two tagging datasets). Experimental results on two source separation datasets, show this approach can produce separation estimates for a wider variety of sources than any tested supervised or unsupervised system. This work points to the vast and heretofore untapped potential of large pretrained music models for audio-to-audio tasks like source separation.