Confidence-Based Self-Training for EMG-to-Speech: Leveraging Synthetic EMG for Robust Modeling

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Xiaodan, Gao, Xiaoxue, Quoy, Mathias, Pitti, Alexandre, Chen, Nancy F.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912813007503360
author Chen, Xiaodan
Gao, Xiaoxue
Quoy, Mathias
Pitti, Alexandre
Chen, Nancy F.
author_facet Chen, Xiaodan
Gao, Xiaoxue
Quoy, Mathias
Pitti, Alexandre
Chen, Nancy F.
contents Voiced Electromyography (EMG)-to-Speech (V-ETS) models reconstruct speech from muscle activity signals, facilitating applications such as neurolaryngologic diagnostics. Despite its potential, the advancement of V-ETS is hindered by a scarcity of paired EMG-speech data. To address this, we propose a novel Confidence-based Multi-Speaker Self-training (CoM2S) approach, along with a newly curated Libri-EMG dataset. This approach leverages synthetic EMG data generated by a pre-trained model, followed by a proposed filtering mechanism based on phoneme-level confidence to enhance the ETS model through the proposed self-training techniques. Experiments demonstrate our method improves phoneme accuracy, reduces phonological confusion, and lowers word error rate, confirming the effectiveness of our CoM2S approach for V-ETS. In support of future research, we will release the codes and the proposed Libri-EMG dataset-an open-access, time-aligned, multi-speaker voiced EMG and speech recordings.
format Preprint
id arxiv_https___arxiv_org_abs_2506_11862
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Confidence-Based Self-Training for EMG-to-Speech: Leveraging Synthetic EMG for Robust Modeling
Chen, Xiaodan
Gao, Xiaoxue
Quoy, Mathias
Pitti, Alexandre
Chen, Nancy F.
Sound
Audio and Speech Processing
Signal Processing
Voiced Electromyography (EMG)-to-Speech (V-ETS) models reconstruct speech from muscle activity signals, facilitating applications such as neurolaryngologic diagnostics. Despite its potential, the advancement of V-ETS is hindered by a scarcity of paired EMG-speech data. To address this, we propose a novel Confidence-based Multi-Speaker Self-training (CoM2S) approach, along with a newly curated Libri-EMG dataset. This approach leverages synthetic EMG data generated by a pre-trained model, followed by a proposed filtering mechanism based on phoneme-level confidence to enhance the ETS model through the proposed self-training techniques. Experiments demonstrate our method improves phoneme accuracy, reduces phonological confusion, and lowers word error rate, confirming the effectiveness of our CoM2S approach for V-ETS. In support of future research, we will release the codes and the proposed Libri-EMG dataset-an open-access, time-aligned, multi-speaker voiced EMG and speech recordings.
title Confidence-Based Self-Training for EMG-to-Speech: Leveraging Synthetic EMG for Robust Modeling
topic Sound
Audio and Speech Processing
Signal Processing
url https://arxiv.org/abs/2506.11862