A STATISTICAL CAUSAL MODEL FOR THE ASSESSMENT OF DYSARTHRIC SPEECH AND THE UTILITY OF COMPUTER-BASED SPEECH RECOGNITION

The evaluation of the degree of speech impairment and the utility of c omputer recognition of impaired speech are separately and independentl y performed. Particular attention is paid to the question concerning w hether or not there is a relationship between naive listeners' subject ive judgments of impaired speech and the performance of a laboratory v ersion of a speech recognition system. It is a difficult task to relat e a speech impairment rating with speech recognition accuracy. Towards this end, a statistical causal model is proposed. This model is very appealing in its structure to support inference, and thus can be appli ed to perform various assessments such as the success of automatic rec ognition of dysarthric speech. The application of this model is illust rated with a case study of a dysarthric speaker compared against a nor mal speaker serving as a control.