Sound classification system for hearing aids

Abstract

A hearing aid includes a sound classification module to classify environmental sound sensed by a microphone. The sound classification module executes an advanced sound classification algorithm. The hearing aid then processes the sound according to the classification.

1 . A hearing aid, comprising:
 a microphone configured to sense an environmental sound; and   a processor including a sound classification module coupled to the microphone and configured to classify the sound by executing an advanced classification algorithm, and processor configured to process the sound using an outcome of the classification of the sound for specified hearing assistance functions.   
     
     
         2 . The hearing aid of  claim 1 , wherein the sound classification nodule is configured to classify the sound by executing an advanced classification algorithm including a Hidden Markov Model (HMM), 
     
     
         3 . The hearing aid of  claim 1 , wherein the sound classification nodule is configured to classify the sound by executing an advanced classification algorithm including a Gaussian Mixture Model (GMM). 
     
     
         4 . The hearing aid of  claim 1 , wherein the sound classification nodule is configured to classify the sound as one of music, speech, and non-speech. 
     
     
         5 . The hearing aid of  claim 4 , wherein the sound classification nodule is configured to further classify the sound as one of machine noise, wind noise, and other sounds in response to the sound being classified as the non-speech. 
     
     
         6 . The hearing aid of  claim 1 , wherein the sound classification module comprises a feature set having an amount of features for sound classification determined based on performance and computational cost of the sound classification. 
     
     
         7 . The hearing aid of  claim 6 , wherein the feature set comprises 5 to 7 features. 
     
     
         8 . The hearing aid of  claim 7 , wherein the feature set comprises low level features. 
     
     
         9 . The hearing aid of  claim 7 , wherein the feature set comprises Mel-scale Frequency cepstral coefficients (MFCC). 
     
     
         10 . A method for operating a hearing aid, comprising:
 sensing an environmental sound;   classifying the sound by executing an advanced classification algorithm; and   processing the sound using an outcome of the classification of the sound for specified hearing assistance functions.   
     
     
         11 . The method of  claim 10 , wherein classifying the sound comprises classifying the sound by using a Hidden Markov Model (HMM), 
     
     
         12 . The method of  claim 10 , wherein classifying the sound comprises classifying the sound by using a Gaussian Mixture Model (GMM). 
     
     
         13 . The method of  claim 10 , wherein classifying the sound comprises classifying the sound using a two-stage environment classification scheme. 
     
     
         14 . The method of  claim 13 , wherein using the two-stage environment classification scheme comprises classifying the sound as one of music, speech or non-speech. 
     
     
         15 . The method of  claim 14 , wherein using the two-stage environment classification scheme further comprises classifying the sound as one of machine noise, wind noise, and other sounds in response to the sound being classified as the non-speech. 
     
     
         16 . The method  claim 10 , comprising selecting the advanced classification algorithm based on based on performance and computational cost for the classifying the sound. 
     
     
         17 . The method of  claim 16 , further comprising selecting a feature set for the classifying the sound based on the performance and computational cost for the classifying the sound. 
     
     
         18 . The method of  claim 17 , further comprising selecting an amount of features in the feature set based on the performance and computational cost for the classifying the sound. 
     
     
         19 . The method of  claim 18 , wherein selecting the amount of features in the feature set comprises selecting 5 to 7 features. 
     
     
         20 . The method of  claim 19 , wherein selecting the amount of features in the feature set comprises selecting low level features.