Interpretable Latent Space Using Space-Filling Curves for Phonetic Analysis in Voice Conversion

Loading...
Thumbnail Image

Access rights

openAccess

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Date

2023

Major/Subject

Mcode

Degree programme

Language

en

Pages

5
306-310

Series

Proceedings of Interspeech Conference, Volume 2023-August, Interspeech

Abstract

Vector quantized variational autoencoders (VQ-VAE) are well-known deep generative models, which map input data to a latent space that is used for data generation. Such latent spaces are unstructured and can thus be difficult to interpret. Some earlier approaches have introduced a structure to the latent space through supervised learning by defining data labels as latent variables. In contrast, we propose an unsupervised technique incorporating space-filling curves into vector quantization (VQ), which yields an arranged form of latent vectors such that adjacent elements in the VQ codebook refer to similar content. We applied this technique to the latent codebook vectors of a VQ-VAE, which encode the phonetic information of a speech signal in a voice conversion task. Our experiments show there is a clear arrangement in latent vectors representing speech phones, which clarifies what phone each latent vector corresponds to and facilitates other detailed interpretations of latent vectors.

Description

Keywords

Interpretable latent space, phonetic analysis, space-filling curves, vector quantization, voice conversion

Other note

Citation

Vali, M & Bäckström, T 2023, Interpretable Latent Space Using Space-Filling Curves for Phonetic Analysis in Voice Conversion . in Proceedings of Interspeech Conference . vol. 2023-August, Interspeech, International Speech Communication Association (ISCA), pp. 306-310, Interspeech, Dublin, Ireland, 20/08/2023 . https://doi.org/10.21437/Interspeech.2023-1549