A channel adaptive wide dynamic range compression method for digital hearing aids

Abstract

The invention discloses a channel-adaptive digital hearing aid wide dynamic range compression method. First, an asymmetric filter bank decomposition and synthesis algorithm is selected by simulating the hearing characteristics of the human ear, and then the filter is designed according to the patient's audiogram and a psychoacoustic model. The number of channels and parameters of the filter group can be used to obtain a personalized filter bank that meets the hearing loss of the patient; finally, the digital hearing aid using the above method performs wide dynamic range compression on the input sound signal sub-channels; the specific steps of compression include: automatically compressing the input signal The adaptive channel filter bank is decomposed to obtain the signal of the adaptive channel; the loudness compensation is performed on each channel signal after decomposition; the filter bank synthesis is performed on the compensated channel signal to obtain a full-band signal; the integrated full-band signal is The signal is converted to an audio signal output. While meeting the performance requirements, the computational complexity of the system is reduced, and the patient's speech intelligibility is improved.

Description

technical field [0001] The invention relates to the technical field of acoustic signal processing of digital hearing aids, in particular to a channel-adaptive wide dynamic range compression method for digital hearing aids. Background technique [0002] The main function of digital hearing aids is loudness compensation, which is to formulate a personalized compensation plan that is most suitable for the patient according to the patient's hearing loss. It involves two problems: (1) how to compensate according to the hearing loss of the patient at different frequencies; (2) how to compensate according to the sound intensity of the input sound, so as to ensure that the patient can clearly understand the weak signal Hear it, and make sure not to make noise when the signal is strong. [0003] The specific implementation method of loudness compensation is wide dynamic range compression (WDRC). WDRC decomposes the sound signal into multiple channel signals, and the gain of each ch...

AI Technical Summary

Problems solved by technology

It involves two issues: (1) how to compensate according to the patient's hearing loss at different frequency points; (2) how to compensate according to the sound intensity of the input sound, so as to ensure that the patient can clearly hear the signal when the signal is weak Hear it, but also ensure that it is not noisy when the signal is strong

First, filter banks with a fixed number of channels cannot effectively compensate for the complex hearing loss of hearing-impaired patients

If the patient has different hearing losses at adjacent frequency points, and these frequency points are located in the same sub-band, different compensation values ​​cannot be given for hearing loss at different frequency points

Second, it wastes precious computation

Some patients have no hearing loss in a wide frequency band, or have the same hearing loss. The algorithm still needs to divide different channels for separate calculations, wasting valuable calculations and power consumption of digital hearing aids

Images