Enhancement of continuous dysarthric speech

Dysarthria is a set of motor-speech disorders resulting due to neurological injuries. It affects the motor component of the motor-speech system. Disruption in muscular control makes the speech imperfect. As a result, dys-arthric speech is not as comprehensible as normal speech. Most often, people with dysarthria have problems with communicating and this inhibits their social participation. Hence, for effective communication, it is extremely vital that we develop assistive speech technologies for people with dysarthria. The aim is to improve the naturalness and intelligibility of dysarthric speech while retaining the characteristics of the speaker. For this purpose, durational attributes across dysarthric and normal speech utterances are first studied. An automatic technique is developed to correct dysarthric speech to bring it closer to normal speech. The performance of this technique is compared with that of two other techniques available in the literature - a formant resynthesis technique and a hidden Markov model based adaptive speech synthesis technique. Subjective evaluations show a preference for dysarthric speech modified using the proposed approach over existing approaches.