Title :
A Single-channel Loudness Compensation Method for Digital Hearing Aids Based on Multi-resolution Wavelet
Author :
Lin Zhimin ; Zhang Linghua ; Yang Zi
Author_Institution :
Coll. of Telecommun. & Inf. Eng., Nanjing Univ. of Post & Telecommun., Nanjing, China
Abstract :
The conventional multi-channel loudness compensation algorithm used in digital hearing aid ignores the protection of speech feature, which results in the deformation of speech structure and low recognition rate. In this paper, a single-channel method based on multi-resolution wavelet is proposed. Multi-resolution wavelet is involved in decomposing and reconstructing the signal, so that the envelope of signal spectrum can be extracted. Then, calculate the required gain of each feature point. After that, the gain of entire signal spectrum is obtained by interpolation algorithm. Experiments show that the proposed algorithm can not only compensate the loudness of speech, but also effectively protect the speech feature, which improves the speech recognition rate.
Keywords :
hearing aids; interpolation; signal reconstruction; speech recognition; wavelet transforms; digital hearing aids; interpolation algorithm; multiresolution wavelet; signal spectrum; single channel loudness compensation method; speech feature; speech loudness; speech recognition; Algorithm design and analysis; Auditory system; Feature extraction; Speech; Speech recognition; Wavelet transforms; Interpolation Algorithm; Loudness Compensation; Multiresolution Analysis; Spectral Envelope; Speech Feature;
Conference_Titel :
Instrumentation, Measurement, Computer, Communication and Control (IMCCC), 2012 Second International Conference on
Conference_Location :
Harbin
Print_ISBN :
978-1-4673-5034-1
DOI :
10.1109/IMCCC.2012.376
