In this paper, we propose a new speech enhancement algorithm based on wavelet packet decomposition and mask filtering. In the traditional mask filtering such as ideal binary mask (IBM), the basic idea is to classify speech components as target signal and non-speech components as background noises. However, speech and non-speech components cannot be well separated in target signal and background noise. Therefore, the IBM has residual noise and signal loss. To overcome this problem, the proposed algorithm used semi-soft mask filter to exponentially increase. The semi-soft mask minimizes signal loss and the exponential filter removes residual noise. We performed experiments using various types of speech and noise signals, and experimental results show that the proposed algorithm achieves better performances than the traditional other speech enhancement algorithms.

This paper presents a voice activity detection (VAD) approach using a perceptual wavelet entropy neighbor slope (PWENS) in a low signal-to-noise (SNR) environment and with a variety of noise types. The basis for our study is to use acoustic features that have large entropy variance for each wavelet critical band. The speech signal is decomposed by the proposed perceptual wavelet packet decomposition (PWPD), and the VAD function is extracted by PWENS. Finally, VAD is decided by the proposed VAD decision rule using two memory buffers. In order to evaluate the performance of the VAD decision, many speech samples and a variety of SNR conditions were used in the experiment. The performance of the VAD decision is confirmed using objective indexes such as a graph of the VAD decision and the relative error rate.

SUMMARYBecause wavelet transforms have the characteristic of decomposing signals that are similar to the human acoustic system, speech enhancement algorithms that are based on wavelet shrinkage are widely used. In this paper, we propose a new speech enhancement algorithm of hearing aids based on wavelet shrinkage. The algorithm has multi-band threshold value and a new wavelet shrinkage function for recursive noise reduction. We performed experiments using various types of authorized speech and noise signals, and our results show that the proposed algorithm achieves significantly better performances compared with other recently proposed speech enhancement algorithms using wavelet shrinkage. key words: speech enhancement, wavelet shrinkage, recursive algorithm, hearing aids

No abstract

Hearing loss in people is increasing because of a rise in the usage of wireless audio multimedia devices. Hearing aids are used as representative hearing rehabilitation devices. Bone conduction hearing aids are recommended for problems in the eardrum and middle ear. Bone conduction is classified according to the driving method into two types, electromagnetic and piezoelectric. Electromagnetic bone conduction causes skin disease and aesthetic problems due to transplantation, high power consumption, and external interference. Piezoelectric bone conduction converts electrical energy into mechanical vibrations, and the characteristics change linearly with size. However, the driving force of ear canal insertion of the piezoelectric body is limited because of the ear canal anatomy. In this paper, a piezoelectric actuator with a bridge structure inserted into the ear canal is proposed. The proposed method is that the displacement amplification ratio was derived using the formula of a bridge-type structure, and the displacement and resonance frequency were derived by finite element analysis (FEA) using different variables. The piezoelectric actuator was fabricated on the basis of FEA simulation results and verified through an artificial mastoid for stimulation in the ear canal. It is expected that the proposed piezoelectric actuator can be used in the various fields for sound and precision control.