Spectro-temporal Deep Features For Disordered Speech Assessment And Recognition
2022 · Mengzhe Geng, Shansong Liu, Jianwei Yu, et al.
Abstract
Automatic recognition of disordered speech remains a highly challenging task to date. Sources of variability commonly found in normal speech including accent, age or gender, when further compounded with the underlying causes of speech impairment and varying severity levels, create large diversity among speakers. To this end, speaker adaptation techniques play a vital role in current speech recognition systems. Motivated by the spectro-temporal level differences between disordered and normal speech that systematically manifest in articulatory imprecision, decreased volume and clarity, slower speaking rates and increased dysfluencies, novel spectro-temporal subspace basis embedding deep features derived by SVD decomposition of speech spectrum are proposed to facilitate both accurate speech intelligibility assessment and auxiliary feature based speaker adaptation of state-of-the-art hybrid DNN and end-to-end disordered speech recognition systems. Experiments conducted on the UASpeech corp
Authors
(none)
Tags
Stats
Related papers
- Speaker Adaptation Using Spectro-temporal Deep Features For Dysarthric And Elderly Speech Recognition (2022)12.02
- Adversarial Data Augmentation Using VAE-GAN For Disordered Speech Recognition (2022)0.00
- Exploiting Cross Domain Acoustic-to-articulatory Inverted Features For Disordered Speech Recognition (2022)8.09
- On-the-fly Feature Based Rapid Speaker Adaptation For Dysarthric And Elderly Speech Recognition (2022)6.34
- Weak-supervised Dysarthria-invariant Features For Spoken Language Understanding Using An FHVAE And Adversarial Training (2022)2.26
- Homogeneous Speaker Features For On-the-fly Dysarthric And Elderly Speaker Adaptation (2024)0.00
- Personalized Adversarial Data Augmentation For Dysarthric And Elderly Speech Recognition (2022)11.49
- Robust Vocal Quality Feature Embeddings For Dysphonic Voice Detection (2022)7.16