Robust automatic intelligibility assessment techniques evaluated on speakers treated for head and neck cancer

Authors:
Catherine Middag;Renee Clapham;Rob Van Son;Jean-Pierre Martens
Affiliations:
MultiMediaLab, ELIS, UGent, Sint-Pietersnieuwstraat 41, 9000 Ghent, Belgium;Amsterdam Center for Language and Communication, University of Amsterdam, Spuistraat 210, 1012 VT Amsterdam, The Netherlands and The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterda ...;The Netherlands Cancer Institute, Plesmanlaan 121, 1066 CX Amsterdam, The Netherlands and Amsterdam Center for Language and Communication, University of Amsterdam, Spuistraat 210, 1012 VT Amsterda ...;MultiMediaLab, ELIS, UGent, Sint-Pietersnieuwstraat 41, 9000 Ghent, Belgium
Venue:
Computer Speech and Language
Year:
2014

Citing 5
Cited 1

Bagging predictors

Machine Learning
Disordered speech assessment using automatic methods based on quantitative measures

EURASIP Journal on Applied Signal Processing
PEAKS - A system for the automatic evaluation of voice and speech disorders

Speech Communication
Automated intelligibility assessment of pathological speech using phonological features

EURASIP Journal on Advances in Signal Processing - Special issue on analysis and signal processing of oesophageal and pathological voices
Characterization of atypical vocal source excitation, temporal dynamics and prosody for objective measurement of dysarthric word intelligibility

Speech Communication

Developing automatic articulation, phonation and accent assessment techniques for speakers treated for advanced head and neck cancer

Speech Communication

Quantified Score

Hi-index	0.00

Visualization

Abstract

It is generally acknowledged that an unbiased and objective assessment of the communication deficiency caused by a speech disorder calls for automatic speech processing tools. In this paper, a new automatic intelligibility assessment method is presented. The method can predict running speech intelligibility in a way that is robust against changes in the text and against differences in the accent of the speaker. It is evaluated on a Dutch corpus comprising longitudinal data of several speakers who have been treated for cancer of the head and the neck. The results show that the method is as accurate as a human listener in detecting trends in the intelligibility over time. By evaluating the intelligibility predictions made with different models trained on distinct texts and accented speech data, evidence for the robustness of the method against text and accent factors is offered.