Facetron: A Multi-speaker Face-to-Speech Model based on Cross-modal Latent Representations

TL;DR

This paper introduces Facetron, a multi-speaker face-to-speech model that converts face images into speech waveforms using cross-modal latent representations, capable of handling unseen speakers with high quality.

Contribution

The paper presents a novel end-to-end GAN-based face-to-speech model utilizing independent linguistic and speaker features for flexible multi-speaker synthesis, including unseen speakers.

Findings

Outperforms conventional methods in objective and subjective evaluations.

Achieves high accuracy in lip-reading-based linguistic feature extraction.

Produces natural-sounding speech with high speaker and gender similarity.

Abstract

In this paper, we propose a multi-speaker face-to-speech waveform generation model that also works for unseen speaker conditions. Using a generative adversarial network (GAN) with linguistic and speaker characteristic features as auxiliary conditions, our method directly converts face images into speech waveforms under an end-to-end training framework. The linguistic features are extracted from lip movements using a lip-reading model, and the speaker characteristic features are predicted from face images using cross-modal learning with a pre-trained acoustic model. Since these two features are uncorrelated and controlled independently, we can flexibly synthesize speech waveforms whose speaker characteristics vary depending on the input face images. We show the superiority of our proposed model over conventional methods in terms of objective and subjective evaluation results. Specifically, we evaluate the performances of linguistic features by measuring their accuracy on an automatic speech recognition task. In addition, we estimate speaker and gender similarity for multi-speaker and unseen conditions, respectively. We also evaluate the aturalness of the synthesized speech waveforms using a mean opinion score (MOS) test and non-intrusive objective speech quality assessment (NISQA).The demo samples of the proposed and other models are available at https://sam-0927.github.io/