Toward a quantitative basis for assessment and diagnosis of apraxia of speech.

PURPOSE: We explored the reliability and validity of 2 quantitative approaches to document presence and severity of speech properties associated with apraxia of speech (AOS). METHOD: A motor speech evaluation was administered to 39 individuals with aphasia. Audio-recordings of the evaluation were presented to 3 experienced clinicians to determine AOS diagnosis and to rate severity of 11 speech dimensions. Additionally, research assistants coded 11 operationalized metrics of articulation, fluency, and prosody in the same speech samples and in recordings from 20 neurologically healthy participants. RESULTS: Agreement among the 3 clinicians was limited for both AOS diagnosis and perceptual scaling, but inter-observer reliability for the operationalized metrics was strong. The relationships between most operationalized metrics and mean severity ratings for corresponding perceptual dimensions were moderately strong and statistically significant. Both perceptual scaling and operationalized quantification approaches were sensitive to the presence or absence of AOS. CONCLUSIONS: Perceptual scaling and operationalized metrics are promising quantification techniques that can help establish diagnostic transparency for AOS. However, because satisfactory reliability cannot be assumed for scaling techniques, effective training and calibration procedures should be implemented. Operationalized metrics show strong potential for enhancing diagnostic objectivity and sensitivity.

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 computer recognition of impaired speech are separately and independently performed. Particular attention is paid to the question concerning whether or not there is a relationship between naive listeners' subjective judgements of impaired speech and the performance of a laboratory version of a speech recognition system. It is a difficult task to relate 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 applied to perform various assessments such as the success of automatic recognition of dysarthric speech. The application of this model is illustrated with a case study of a dysarthric speaker compared against a normal speaker serving as a control.