On-the-Fly Feature Based Rapid Speaker Adaptation for Dysarthric and Elderly Speech Recognition
Preprint
Abstract
Accurate recognition of dysarthric and elderly speech remain challenging tasks to date. Speaker-level heterogeneity attributed to accent or gender, when aggregated with age and speech impairment, create large diversity among these speakers. Scarcity of speaker-level data limits the practical use of data-intensive model based speaker adaptation methods. To this end, this paper proposes two novel forms of data-efficient, feature-based on-the-fly speaker adaptation methods: variance-regularized spectral basis embedding (SVR) and spectral feature driven f-LHUC transforms. Experiments conducted on UASpeech dysarthric and DementiaBank Pitt elderly speech corpora suggest the proposed on-the-fly speaker adaptation approaches consistently outperform baseline iVector adapted hybrid DNN/TDNN and E2E Conformer systems by statistically significant WER reduction of 2.48%-2.85% absolute (7.92%-8.06% relative), and offline model based LHUC adaptation by 1.82% absolute (5.63% relative) respectively.Keywords
All Related Versions
This publication has 47 references indexed in Scilit:
- Phonetic Analysis of Dysarthric Speech Tempo and Applications to Robust Personalised Dysarthric Speech Recognition2019
- An investigation of augmenting speaker representations to improve speaker normalisation for DNN-based speech recognition2015
- Speaker adaptation of neural network acoustic models using i-vectors2013
- Fast speaker adaptation of hybrid NN/HMM model for speech recognition based on discriminative learning of speaker code2013
- Feature engineering in Context-Dependent Deep Neural Networks for conversational speech transcription2011
- Adapting acoustic and lexical models to dysarthric speech2011
- Tools for the analysis of benchmark speech recognition tests2002
- Speaker adaptation using constrained estimation of Gaussian mixturesIEEE Transactions on Speech and Audio Processing, 1995
- Subspace-based signal analysis using singular value decompositionProceedings of the IEEE, 1993
- Temporal Characteristics of the Speech of Normal Elderly AdultsJournal of Speech, Language, and Hearing Research, 1987