# SPADE: Self-supervised Pretraining for Acoustic DisEntanglement

arXiv: https://arxiv.org/abs/2302.01483 · HTML: https://mlsanity.com/paper/2302.01483/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 20 references that are in the map.
- First posted on arXiv 3 Feb 2023.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [AUTOVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss](https://mlsanity.com/paper/1905.05879/) (2019)
- [Unsupervised Speech Decomposition via Triple Information Bottleneck](https://mlsanity.com/paper/2004.11284/) (2020)
- [ContentVec: An Improved Self-Supervised Speech Representation by Disentangling Speakers](https://mlsanity.com/paper/2204.09224/) (2022)
- [Disentangled Speech Embeddings using Cross-modal Self-supervision](https://mlsanity.com/paper/2002.08742/) (2020)
- [Robust Disentangled Variational Speech Representation Learning for Zero-shot Voice Conversion](https://mlsanity.com/paper/2203.16705/) (2022)
- [Privacy-preserving Voice Analysis via Disentangled Representations](https://mlsanity.com/paper/2007.15064/) (2020)
- [Adversarially learning disentangled speech representations for robust multi-factor voice conversion](https://mlsanity.com/paper/2102.00184/) (2021)
- [Unsupervised Adaptation with Interpretable Disentangled Representations for Distant Conversational Speech Recognition](https://mlsanity.com/paper/1806.04872/) (2018)
- [End-to-end Alexa Device Arbitration](https://mlsanity.com/paper/2112.04914/) (2021)

## Abstract (arXiv)

Self-supervised representation learning approaches have grown in popularity due to the ability to train models on large amounts of unlabeled data and have demonstrated success in diverse fields such as natural language processing, computer vision, and speech. Previous self-supervised work in the speech domain has disentangled multiple attributes of speech such as linguistic content, speaker identity, and rhythm. In this work, we introduce a self-supervised approach to disentangle room acoustics from speech and use the acoustic representation on the downstream task of device arbitration. Our results demonstrate that our proposed approach significantly improves performance over a baseline when labeled training data is scarce, indicating that our pretraining scheme learns to encode room acoustic information while remaining invariant to other attributes of the speech signal.