# Mono-to-stereo through parametric stereo generation

arXiv: https://arxiv.org/abs/2306.14647 · HTML: https://mlsanity.com/paper/2306.14647/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 28 references that are in the map.
- First posted on arXiv 26 Jun 2023.

## Builds on (most cited first, 7 of 7)

- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [MaskGIT: Masked Generative Image Transformer](https://mlsanity.com/paper/2202.04200/) (2022)
- [Muse: Text-To-Image Generation via Masked Generative Transformers](https://mlsanity.com/paper/2301.00704/) (2023)
- [PixelSNAIL: An Improved Autoregressive Generative Model](https://mlsanity.com/paper/1712.09763/) (2017)
- [Upsampling artifacts in neural audio synthesis](https://mlsanity.com/paper/2010.14356/) (2020)
- [Automatic multitrack mixing with a differentiable mixing console of neural audio effects](https://mlsanity.com/paper/2010.10291/) (2020)

## Built on by (most cited first, 3 of 3)

- [MaskSR: Masked Language Model for Full-band Speech Restoration](https://mlsanity.com/paper/2406.02092/) (2024)
- [AV-Surf: Surface-Enhanced Geometry-Aware Novel-View Acoustic Synthesis](https://mlsanity.com/paper/2503.12806/) (2025)
- [StereoSync: Spatially-Aware Stereo Audio Generation from Video](https://mlsanity.com/paper/2510.05828/) (2025)

## Abstract (arXiv)

Generating a stereophonic presentation from a monophonic audio signal is a challenging open task, especially if the goal is to obtain a realistic spatial imaging with a specific panning of sound elements. In this work, we propose to convert mono to stereo by means of predicting parametric stereo (PS) parameters using both nearest neighbor and deep network approaches. In combination with PS, we also propose to model the task with generative approaches, allowing to synthesize multiple and equally-plausible stereo renditions from the same mono signal. To achieve this, we consider both autoregressive and masked token modelling approaches. We provide evidence that the proposed PS-based models outperform a competitive classical decorrelation baseline and that, within a PS prediction framework, modern generative models outshine equivalent non-generative counterparts. Overall, our work positions both PS and generative modelling as strong and appealing methodologies for mono-to-stereo upmixing. A discussion of the limitations of these approaches is also provided.