Individually tailored spectral-change enhancement for the hearing impaired

Jing Chen, Brian C.J. Moore, Thomas Baer, Xihong Wu

科研成果: 期刊稿件文章同行评审

5 引用 (Scopus)

摘要

An algorithm for enhancing spectral changes over time was previously shown to improve the intelligibility of speech in steady speech-spectrum noise (SSN) for hearing-impaired subjects but tended to impair intelligibility for speech in a background of two-talker speech. Large individual differences were found and the application of a genetic algorithm for selecting the "best" parameter values for each listener was found to be beneficial. In the present study, the spectral-change enhancement (SCE) processing was modified by individually tailoring the degree of SCE based on the frequency-dependent hearing loss of the subjects, and by using finer frequency resolution. The effect of the modified SCE processing on the intelligibility and quality of speech in SSN and babble noise (BBN) was evaluated. Ten subjects with mild to moderate hearing loss were tested twice for all tests. The SCE processing led to small but significant improvements in the intelligibility of speech in both SSN and BBN, while the effect of the SCE processing on speech quality was small.

源语言英语
页(从-至)1128-1137
页数10
期刊Journal of the Acoustical Society of America
143
2
DOI
出版状态已出版 - 1 2月 2018
已对外发布

指纹

探究 'Individually tailored spectral-change enhancement for the hearing impaired' 的科研主题。它们共同构成独一无二的指纹。

引用此