# Automatic multitrack mixing with a differentiable mixing console of neural audio effects

arXiv: https://arxiv.org/abs/2010.10291 · HTML: https://mlsanity.com/paper/2010.10291/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Oct 2020.

## Built on by (most cited first, 27 of 27)

- [Fast Timing-Conditioned Latent Audio Diffusion](https://mlsanity.com/paper/2402.04825/) (2024)
- [ETTA: Elucidating the Design Space of Text-to-Audio Models](https://mlsanity.com/paper/2412.19351/) (2024)
- [Audio Super-Resolution with Latent Bridge Models](https://mlsanity.com/paper/2509.17609/) (2025)
- [High Fidelity Text-Guided Music Editing via Single-Stage Flow Matching](https://mlsanity.com/paper/2407.03648/) (2024)
- [Automatic music mixing with deep learning and out-of-domain data](https://mlsanity.com/paper/2208.11428/) (2022)
- [Modelling black-box audio effects with time-varying feature modulation](https://mlsanity.com/paper/2211.00497/) (2022)
- [UALM: Unified Audio Language Model for Understanding, Generation and Reasoning](https://mlsanity.com/paper/2510.12000/) (2025)
- [DiffMoog: a Differentiable Modular Synthesizer for Sound Matching](https://mlsanity.com/paper/2401.12570/) (2024)
- [End-to-end Music Remastering System Using Self-supervised and Adversarial Training](https://mlsanity.com/paper/2202.08520/) (2022)
- [Music Foundation Model as Generic Booster for Music Downstream Tasks](https://mlsanity.com/paper/2411.01135/) (2024)
- [SonicMaster: Towards Controllable All-in-One Music Restoration and Mastering](https://mlsanity.com/paper/2508.03448/) (2025)
- [SoundReactor: Frame-level Online Video-to-Audio Generation](https://mlsanity.com/paper/2510.02110/) (2025)
- [Distortion Audio Effects: Learning How to Recover the Clean Signal](https://mlsanity.com/paper/2202.01664/) (2022)
- [Adoption of AI Technology in the Music Mixing Workflow: An Investigation](https://mlsanity.com/paper/2304.03407/) (2023)
- [Automatic DJ Transitions with Differentiable Audio Effects and Generative Adversarial Networks](https://mlsanity.com/paper/2110.06525/) (2021)
- [Mono-to-stereo through parametric stereo generation](https://mlsanity.com/paper/2306.14647/) (2023)
- [AMSS-Net: Audio Manipulation on User-Specified Sources with Textual Queries](https://mlsanity.com/paper/2104.13553/) (2021)
- [Modeling Analog Dynamic Range Compressors using Deep Learning and State-space Models](https://mlsanity.com/paper/2403.16331/) (2024)
- [Prevailing Research Areas for Music AI in the Era of Foundation Models](https://mlsanity.com/paper/2409.09378/) (2024)
- [Evaluating Neural Networks Architectures for Spring Reverb Modelling](https://mlsanity.com/paper/2409.04953/) (2024)
- [Compression of Higher Order Ambisonics with Multichannel RVQGAN](https://mlsanity.com/paper/2411.12008/) (2024)
- [Improving Inference-Time Optimisation for Vocal Effects Style Transfer with a Gaussian Prior](https://mlsanity.com/paper/2505.11315/) (2025)
- [WildFX: A DAW-Powered Pipeline for In-the-Wild Audio FX Graph Modeling](https://mlsanity.com/paper/2507.10534/) (2025)
- [TimberAgent: Gram-Guided Retrieval for Executable Music Effect Control](https://mlsanity.com/paper/2603.09332/) (2026)
- [It`s All About Speed: AI`s Impact on Workflow in Music Production](https://mlsanity.com/paper/2605.29931/) (2026)
- [Decomposer: Learning to Decompile Symbolic Music to Programs](https://mlsanity.com/paper/2607.01849/) (2026)
- [Diff2Mix: Controllable Music Mixing via Diffusion Models and Differentiable Audio Effects](https://mlsanity.com/paper/2608.05442/) (2026)

## Abstract (arXiv)

Applications of deep learning to automatic multitrack mixing are largely unexplored. This is partly due to the limited available data, coupled with the fact that such data is relatively unstructured and variable. To address these challenges, we propose a domain-inspired model with a strong inductive bias for the mixing task. We achieve this with the application of pre-trained sub-networks and weight sharing, as well as with a sum/difference stereo loss function. The proposed model can be trained with a limited number of examples, is permutation invariant with respect to the input ordering, and places no limit on the number of input sources. Furthermore, it produces human-readable mixing parameters, allowing users to manually adjust or refine the generated mix. Results from a perceptual evaluation involving audio engineers indicate that our approach generates mixes that outperform baseline approaches. To the best of our knowledge, this work demonstrates the first approach in learning multitrack mixing conventions from real-world data at the waveform level, without knowledge of the underlying mixing parameters.