[Response of the auditory nerve to pure sounds having abrupt onset].
Explore the source record for details and available documents.
Biomedical subjects
Publications and source records attributed to B Delgutte.
Explore the source record for details and available documents.
Two experimental techniques were used in this study: (1) an analysis of F0 contours of French sentences containing words likely to express the speaker's attitude (modals, negative, adverbs, quantifiers) and (2) listeners' judgements of the naturalness of synthetic sentences in which the F0 contour was systematically manipulated. The F0 contours were characterized by schematic patterns [J. Vaissière Res. Lab. Electron., Q. Prog. Report No. 115, MIT, 251-262 (1975)]. For all the sentences that were studied, the configurations of patterns frequently observed in the analyzed corpus were judged most natural when synthesized. This result suggests that the features of the F0 contours most relevant to naturalness are preserved in the pattern representation. Three types of patterns were distinguished: (1) a prominent, demarcative pattern (P2); (2) A PROMINENT, NOndemarcative pattern (PE); (3) nonprominent, nondemarcative patterns, A P2 pattern often occurred at the end of a syntactic constituent. Listeners' judgements showed that a P2 at the end of a syntactic constituent is rarely required for naturalness, but a P2 in contradiction to syntax causes unnaturalness. Prominent patterns (P2 or PE) often occurred on negatives, adverbs, and quantifiers. However, the presence or absence of a prominent pattern on these words does not affect naturalness.
The usual premise underlying developmental work on cochlear prostheses as prospective cures for profound deafness is that the auditory nerve can be electrically stimulated in such a manner that communicative skills can be developed or maintained. Physiologic recordings from single fibers in the cat's auditory nerve and attempts to model these responses have generated a description of how the auditory nerve codes complex sounds such as speech. This work suggests that certain minimal cues might have to be present at the level of the auditory nerve in order that adequate discrimination of specific speech signals can take place. The prospects for achieving a useful prosthesis in the near future will be evaluated in terms of what can be expected from current attempts to code the artificial stimulation properly.