A Two Microphone-Based Approach for Detecting and Identifying Speech Sounds in Hearing Support System

2018 ◽  
Vol 2 (2) ◽  
pp. 74-86
Author(s):  
Andre Sitompul ◽  
Masafumi Nishimura
Keyword(s):  
Support Vector Machine ◽  
Support System ◽  
Hearing Impaired ◽  
Support Vector ◽  
Speech Sounds ◽  
Sound Sources ◽  
Robust Approach ◽  
Input Channel ◽  
Impaired People ◽  
Life Threatening

For people with hearing disabilities, not only would give them difficulties in going through their everyday lives but also sometimes could be life threatening. In this research we proposed a simple, yet robust approach for helping the hearing-impaired people in identifying the important sounds around them by using two microphones as input channel that could be worn around the person’s head as a substitute for their ears. This device then could be used to record the situation of the surroundings, and then the system would estimate the Direction of Arrival (DOA) of the sound sources, then detect and classify them using Support Vector Machine (SVM) into target speech or noise category. As the results, system’s classifier could produce FAR and FRR as low as 2%, in which 274 out of 280 samples were successfully classified as target speeches and 22 from the total of 27 noise samples were successfully classified as noise

scholarly journals A Two Microphone-Based Approach for Detecting and Identifying Speech Sounds in Hearing Support System

2019 ◽  
Vol 2 (2) ◽  
pp. 74-86
Author(s):  
Andre Sitompul ◽  
Masafumi Nishimura
Keyword(s):  
Support Vector Machine ◽  
Support System ◽  
Hearing Impaired ◽  
Support Vector ◽  
Speech Sounds ◽  
Sound Sources ◽  
Robust Approach ◽  
Input Channel ◽  
Impaired People ◽  
Life Threatening

For people with hearing disabilities, not only would give them difficulties in going through their everyday lives but also sometimes could be life threatening. In this research we proposed a simple, yet robust approach for helping the hearing-impaired people in identifying the important sounds around them by using two microphones as input channel that could be worn around the person’s head as a substitute for their ears. This device then could be used to record the situation of the surroundings, and then the system would estimate the Direction of Arrival (DOA) of the sound sources, then detect and classify them using Support Vector Machine (SVM) into target speech or noise category. As the results, system’s classifier could produce FAR and FRR as low as 2%, in which 274 out of 280 samples were successfully classified as target speeches and 22 from the total of 27 noise samples were successfully classified as noise

Localization of Sound Sources in Normal Hearing and in Hearing Impaired People

2011 ◽  
Vol 36 (4) ◽  
pp. 683-694 ◽  
Author(s):  
Marek Niewiarowicz ◽  
Tomasz Kaczmarek
Keyword(s):  
Traffic Noise ◽  
Normal Hearing ◽  
Hearing Impaired ◽  
Directional Hearing ◽  
Sound Sources ◽  
Hearing Ability ◽  
Modified Method ◽  
Noise Type ◽  
Impaired People ◽  
New System

Abstract This article presents results of investigations of the angle of directional hearing acuity (ADHA) as a measure of the spatial hearing ability with a special emphasis on people with hearing impairments. A modified method proposed by Zakrzewski has been used - ADHA values have been determined for 8 azimuths in the horizontal plane at the height of the listeners' head. The two-alternative-forced-choice method (2AFC), based on a new system of listeners' responses (left - right instead of no difference - difference in location of sound sources) was the procedure used in the experiment. Investigations were carried out for two groups of subjects: normal hearing people (9 persons) and hearing impaired people (sensorineural hearing loss and tinnitus - 9 persons). In the experiment different acoustic signals were used: sinusoidal signals (pure tones), 1/3 octave noise, amplitude modulated 1/3 octave noise, CCITT speech and traffic noises and signals corresponding to personal character of tinnitus for individual subjects. The results obtained in the investigations showed, in general, a better localization of the sound source for noise type signals than those for tonal signals. Inessential differences exist in ADHA values for particular signals between the two groups of subjects. On the other hand, significant differences for tinnitus signals and traffic noise signals were stated. A new system of listeners' responses was used and appeared efficient (less dispersion of results compared to the standard system).

Sign in / Sign up

Export Citation Format

Share Document