Acoustic transformations to improve the intelligibility of dysarthric speech

Authors:
Frank Rudzicz
Affiliations:
University of Toronto, Toronto, Ontario, Canada
Venue:
SLPAT '11 Proceedings of the Second Workshop on Speech and Language Processing for Assistive Technologies
Year:
2011

Citing 11
Cited 0

From text to speech: the MITalk system

From text to speech: the MITalk system
Pitch-synchronous waveform processing techniques for text-to-speech synthesis using diphones

Speech Communication
Schaum's Outline of Digital Signal Processing

Schaum's Outline of Digital Signal Processing
Numerical Recipes in C: The Art of Scientific Computing

Numerical Recipes in C: The Art of Scientific Computing
Intelligent Processing of Stuttered Speech

Journal of Intelligent Information Systems
A speech-first model for repair detection and correction

ACL '93 Proceedings of the 31st annual meeting on Association for Computational Linguistics
Formant tracking linear prediction model using HMMs and Kalman filters for noisy speech processing

Computer Speech and Language
Improving the intelligibility of dysarthric speech

Speech Communication
Rhythm and Transforms

Rhythm and Transforms
Temporally variable multi-aspect auditory morphing enabling extrapolation without objective and perceptual breakdown

ICASSP '09 Proceedings of the 2009 IEEE International Conference on Acoustics, Speech and Signal Processing
The TORGO database of acoustic and articulatory speech from speakers with dysarthria

Language Resources and Evaluation

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes modifications to acoustic speech signals produced by speakers with dysarthria in order to make those utterances more intelligible to typical listeners. These modifications include the correction of tempo, the adjustment of formant frequencies in sonorants, the removal of aberrant voicing, the deletion of phoneme insertion errors, and the replacement of erroneously dropped phonemes. Through simple evaluations of intelligibility with naïve listeners, we show that the correction of phoneme errors results in the greatest increase in intelligibility and is therefore a desirable mechanism for the eventual creation of augmentative application software for individuals with dysarthria.