Shigeko Hatano

Automatic Detection System for Cough Sounds as a Symptom of Abnormal Health Condition

The problem of attending to the health of the aged who live alone has became an important issue in developed countries. One way of solving the problem is to check their health condition by a remote-monitoring technique and support them with well-timed treatment. The purpose of this study is to develop an automatic system that can monitor a health condition in real time using acoustical information and detect an abnormal symptom. In this study, cough sound was chosen as a representative acoustical symptom of abnormal health conditions. For the development of the system distinguishing a cough sound from other environmental sounds, a hybrid model was proposed that consists of an artificial neural network (ANN) model and a hidden Markov model (HMM). The ANN model used energy cepstral coefficients obtained by filter banks based on human auditory characteristics as input parameters representing a spectral feature of a sound signal. Subsequently, an output of this ANN model and a filtered envelope of the signal were used for making an input sequence for the HMM that deals with the temporal variation of the sound signal. Compared with the conventional HMM using Mel-frequency cepstral coefficients, the proposed hybrid model improved recognition rates on low SNR from 5 dB down to -10 dB. Finally, a preliminary prototype of the automatic detection system was simply illustrated.

Energy distribution for sournd quality improvement of exhaust noise of crusier type of motorcycle

The demand for motorcycles has seen steady growth in recent years. For motorcycles, as well as automobiles in general, sound quality (SQ) is one of the important factors affecting riding comfort. Thus, as with the wider field of automobile manufacturing, motorcycle manufacturers regard SQ-based noise control as a useful tool for improving competitiveness and popularity in the global commercial market. Among the numerous parts of a motorcycle, the exhaust system is the key to determining the SQ, as well as its performance. In this study, the distribution of the exhaust noise energy of a cruiser type of motorcycle in the frequency domain was proposed to improve its SQ. To this end, the audible frequency range was divided into three bands based on its effect on the perceptual feeling described as pleasant. To obtain a desirable energy distribution, a subjective listening experiment based on a semantic differential method was conducted with the modified exhaust noises by changing the band level of each division of the frequency band. A hierarchical clustering analysis and factor analysis were applied to the results of the subjective experiment in order to determine the relation between the subjective feeling and the variation in the spectral shapes of the exhaust noises. The results showed that perceptual feelings described as pleasant, expensive, and classic were strongly influenced by a change in the band level in the low-frequency range below 300 Hz, whereas the hard and crispy perceptions were influenced in the high-frequency range above 1 kHz. To quantitatively represent such situations, the energy ratio and partial-loudness ratio representing the energy balance were defined and calculated. Finally, it was found that a larger energy ratio or partial-loudness ratio resulted in a better SQ for the exhaust noise of the cruiser type of motorcycle.

Sound quality design of vehicle noise

In order to improve sound quality of vehicle noise, the idling exterior noise of a small diesel engine truck and the interior noise of a small‐size sedan were varied artificially using a digital signal‐processing technique in the frequency domain. First, the original exterior idling noise and its variations were tested using a paired‐comparison technique concerning their unpleasantness. As a result, the reduction of the high‐frequency component, i.e., −6 dBA compared with the original, was most effective in reducing unpleasantness of exterior idling noise. To realize this effect in a very small diesel engine truck, an engine compartment enclosure was fixed to the original truck. The result was quite successful in achieving a significant reduction in the high‐frequency noise component. Second, interior noises of a small size sedan with four cylinders and a four‐cycle engine were varied in their second‐order component of engine revolutions and their booming sensation was tested using magnitude estimation an...