Dysarthric Speech
Dysarthric speech, characterized by impaired articulation due to neurological conditions, presents a significant challenge for automatic speech recognition (ASR) and related applications. Current research focuses on developing robust ASR systems for dysarthric speech using techniques like self-supervised learning (e.g., HuBERT, wav2vec 2.0), prototype-based adaptation, and generative adversarial networks (GANs) to address data scarcity and inter-speaker variability. These advancements aim to improve speech recognition accuracy, intelligibility assessment, and even severity classification, ultimately enhancing communication and quality of life for individuals with dysarthria.
Papers
Perceiver-Prompt: Flexible Speaker Adaptation in Whisper for Chinese Disordered Speech Recognition
Yicong Jiang, Tianzi Wang, Xurong Xie, Juan Liu, Wei Sun, Nan Yan, Hui Chen, Lan Wang, Xunying Liu, Feng Tian
Enhancing Voice Wake-Up for Dysarthria: Mandarin Dysarthria Speech Corpus Release and Customized System Design
Ming Gao, Hang Chen, Jun Du, Xin Xu, Hongxiao Guo, Hui Bu, Jianxing Yang, Ming Li, Chin-Hui Lee
Training Data Augmentation for Dysarthric Automatic Speech Recognition by Text-to-Dysarthric-Speech Synthesis
Wing-Zin Leung, Mattias Cross, Anton Ragni, Stefan Goetze
CoLM-DSR: Leveraging Neural Codec Language Modeling for Multi-Modal Dysarthric Speech Reconstruction
Xueyuan Chen, Dongchao Yang, Dingdong Wang, Xixin Wu, Zhiyong Wu, Helen Meng
Speaker-Independent Dysarthria Severity Classification using Self-Supervised Transformers and Multi-Task Learning
Lauren Stumpf, Balasundaram Kadirvelu, Sigourney Waibel, A. Aldo Faisal
Inappropriate Pause Detection In Dysarthric Speech Using Large-Scale Speech Recognition
Jeehyun Lee, Yerin Choi, Tae-Jin Song, Myoung-Wan Koo