# Privacy-Preserving Speech Representation Learning using Vector Quantization

arXiv: https://arxiv.org/abs/2203.09518 · HTML: https://mlsanity.com/paper/2203.09518/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 11 of its 22 references that are in the map.
- First posted on arXiv 15 Mar 2022.

## Builds on (most cited first, 11 of 11)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Estimating or Propagating Gradients Through Stochastic Neurons for Conditional Computation](https://mlsanity.com/paper/1308.3432/) (2013)
- [Fast Decoding in Sequence Models using Discrete Latent Variables](https://mlsanity.com/paper/1803.03382/) (2018)
- [Unsupervised speech representation learning using WaveNet autoencoders](https://mlsanity.com/paper/1901.08810/) (2019)
- [A Hybrid Deep Learning Architecture for Privacy-Preserving Mobile Analytics](https://mlsanity.com/paper/1703.02952/) (2017)
- [Large Margin Softmax Loss for Speaker Verification](https://mlsanity.com/paper/1904.03479/) (2019)
- [Privacy-Preserving Adversarial Representation Learning in ASR: Reality or Illusion?](https://mlsanity.com/paper/1911.04913/) (2019)
- [What does a network layer hear? Analyzing hidden representations of end-to-end ASR through speech synthesis](https://mlsanity.com/paper/1911.01102/) (2019)
- [To Reverse the Gradient or Not: An Empirical Comparison of Adversarial and Multi-task Learning in Speech Recognition](https://mlsanity.com/paper/1812.03483/) (2018)
- [Configurable Privacy-Preserving Automatic Speech Recognition](https://mlsanity.com/paper/2104.00766/) (2021)
- [Pkwrap: a PyTorch Package for LF-MMI Training of Acoustic Models](https://mlsanity.com/paper/2010.03466/) (2020)

## Abstract (arXiv)

With the popularity of virtual assistants (e.g., Siri, Alexa), the use of speech recognition is now becoming more and more widespread.However, speech signals contain a lot of sensitive information, such as the speaker's identity, which raises privacy concerns.The presented experiments show that the representations extracted by the deep layers of speech recognition networks contain speaker information.This paper aims to produce an anonymous representation while preserving speech recognition performance.To this end, we propose to use vector quantization to constrain the representation space and induce the network to suppress the speaker identity.The choice of the quantization dictionary size allows to configure the trade-off between utility (speech recognition) and privacy (speaker identity concealment).