Spectral Analysis of Sound Errors in Persons With Apraxia of Speech and Aphasia

Incorrect productions of [] were compared to correct productions of [s] and [] in two subjects with apraxia of speech and aphasia. Frequencies of the peak spectral energy (FPSE) were extracted from power spectrums of monosyllabic words. In addition, in order to verify that [s] and [] could be differentiated in terms of FPSE, we obtained speech samples of correct productions of [s] and [] from two normal male speakers. A perceptual task was also completed by requiring two listeners to categorize recorded productions. Results revealed significant differences among sound categories for Subject 1, but not for Subject 2. That is, Subject 1's incorrect [] productions were acoustically different from his homonymous [s] productions, whereas the productions were not acoustically different for Subject 2. These findings suggest that Subject 1's //errors were not errors of substitution, but would be more appropriately characterized as distortions. Conversely, this spectral analysis did not clarify the nature of Subject 2's incorrect // productions; both substitution and distortion remained possible appropriate descriptors of his sound errors.

Key Words: apraxia, acoustics, aphasia, speech, spectral analysis

CiteULike    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?