Youna Ji

A priori SAP estimator based on the magnitude square coherence for dual-channel microphone system

In this paper, we present a time-frequency (TF)-dependent a priori speech absence probability (SAP) estimator utilizing the magnitude square coherence (MSC) between two microphone signals. It is shown that the normalized SNR can be numerically computed from the MSC by solving a quadratic equation. Based on the fact that the normalized SNR is bounded between 0 and 1, we directly use it for the probability of speech absence in each TF-unit. Since this approach does not require prior statistical knowledge of noise and speech, it is not affected by the performance of the noise PSD estimator. Furthermore, unlike the conventional SNR-based estimator, additional mapping strategy is unnecessary. The algorithm was evaluated using the receiver operating characteristic (ROC) curve and it attained higher correct detection rate at a given false-alarm rate than the conventional algorithms.

Robust noise PSD estimation for binaural hearing aids in time-varying diffuse noise field

In this paper, we present an unsupervised noise PSD estimation algorithm for binaural hearing aids in a time-varying diffuse noise field. It is shown that the noise PSD can be obtained from the eigenvalues of the input covariance matrix together with the noise coherence function effective at low frequencies. To reduce the estimation bias due to fast smoothing, pre- and post-compensation methods are proposed. The proposed algorithm is able to track non-stationary noise PSD without tracking delay or underestimation problems. Its performance is independent of the target speech direction and input SNR. Results of the objective parameter evaluation demonstrate the superiority of the proposed algorithm over conventional techniques.

Improved a priori SAP Estimator in Complex Noisy Environment for Dual Channel Microphone System.

In this paper, a priori speech absence probability (SAP) estimator is proposed for accurately obtaining the speech presence probability (SPP) in a complex noise field. Unlike previous techniques, the proposed estimator considers a complex noise sound field where the target speech is corrupted by a coherent interference with diffuse noise around. The proposed algorithm estimates a priori SAP based on the normalized speech to interference plus diffuse noise ratio (SINR) being expressed in terms of the speech to interference ratio (SIR) and the directional to diffuse noise ratio (DDR). The SIR is obtained from a quadratic equation of the magnitude-squared coherence (MSC) between two microphone signals. A performance comparison with several advanced a priori SAP estimators was conducted in terms of the receiver operating characteristic (ROC) curve. The proposed algorithm attains a correct detection rate at a given falsealarm rate that is higher than those attained by conventional algorithms.

Noise reduction for binaural hearing aids using unsupervised diffuse noise estimator

In this paper a new noise reduction algorithm for binaural hearing aids is proposed. This algorithm is capable of suppressing both nonstationary diffuse noise and unknown directional interferences without distorting the directional cues. For the estimation of the diffuse noise power spectral density (PSD), we utilize the eigenstructure of the 2×2 input covariance matrix, together with a compensation process for preventing the underestimation at low frequencies. The interference PSD is estimated from the target cancelled input signals through the signal prediction. Effectiveness of the proposed algorithm was confirmed according to the computer simulations in terms of noise reduction and binaural cue preservation.

Active control of low-frequency underwater echoes using an array of tile projectors

With the introduction of low-frequency active sonar in anti-submarine warfare, there is a growing need for a novel underwater invisibility device that could replace the existing passive anechoic tiles. In this talk, we describe an experiment on active reduction of underwater echoes at low frequencies using an array of tile-shaped projectors. Each tile projector was designed with the aid of finite-element computations and tested in an acoustic tank for transmit voltage sensitivity and directivity. An array of tile projectors, covering a scale model submarine in a large acoustic tank, were driven by control signals that were intended to produce the impedance match between water and the object. Depending on the frequency of the incident wave, echo reduction as large as 6 dB was achieved. [This work has been supported by the Low Observable Technology Research Center program of Defense Acquisition Program Administration and Agency for Defense Development.]

Partitioned Block Frequency Domain Adaptive Filtering Algorithm for Nonlinear Acoustic Echo Cancellation

This paper proposes a robust nonlinear acoustic echo canceller (NAEC) which is effective for modeling the nonlinearity of a speaker module and the long acoustic echo path within a speech communication environment. The proposed NAEC utilizes a sigmoid pre-processor for modeling the speaker nonlinearity and a partitioned block frequnecy-domain adaptive filter for identifying the acoustic echo path with small delay. Simulation results confirmed that the proposed algorithm achieves excellent performance with much lower computational complexity than the previous NAEC.

Secondary Path Estimation Algorithm Based on Residual Music Canceller for Noise Cancelling Headphone

An active noise control (ANC) algorithm for noise canceling headphone is proposed. In this study, the feedback ANC operated with the filtered-x least mean square algorithm (FxLMS) algorithm is used to attenuate the undesired noise. Also an adaptive residual music canceller (RMC) is proposed for enhancing the accuracy of the reference signal of the feedback ANC. Simulation results show that a high quality of music sound can be consistently achieved in a time-varying secondary path situation.

A feedback active noise control for in-ear headphone with robust on-line secondary path modeling

The active noise control (ANC) technique can be used to improve the comfortness and quality of the music listening with an in-ear headphone. For a compact headphone design, the feedback ANC operated with the filtered-x LMS (FxLMS) algorithm is preferred. In practice, the error in the estimated secondary path can be problematic for the FxLMS algorithm, since it can lead the algorithm to an unstable state, and thus cause degradation of sound quality with the in-ear headphone. In this study, an adaptive residual music canceler (RMC) is proposed for enhancing the accuracy of the reference signal of the feedback ANC. Since RMC is designed to track the bias of the current secondary path estimate, the secondary path is continuously updated by combining the previous estimate of the secondary path with the current weight vector of RMC. In addition, variable step-size schemes are developed for both the control and secondary path estimation filters, which enable the ANC system to adapt quickly and robustly to the va...

Binaural noise suppression based on an unbiased estimator of target PSD in complex noise environments

The conventional target power density spectrum (PSD) estimation methods based on the signal prediction inherently produce a biased target PSD because of irrelevant assumptions for the noisy environment. In this paper, an unbiased target PSD is obtained by removing the effect of diffuse noise on the prediction filter. In addition, by on-line estimation of both the noise PSD and target transfer function ratio (TFR) from the input signals, the proposed algorithm achieves robust noise suppression for an unknown target direction under a fast time-varying noisy environment. Computer simulations demonstrate the effectiveness and superiority of the proposed algorithm over the conventional methods.

Development and Implementation of Noise-Canceling Technology for Digital Stethoscope

In this paper, an algorithm for suppressing acoustic noises contained in stethoscope sound is proposed and implemented in real-time using an embedded DSP system. Sound collected by stethoscope is down-sampled and band-pass filtered, and later an NLMS adaptive filter is used to cancel the acoustic noise induced from external noise sources. Also, the unpredictable impulsive noises due to fabric friction and instantaneous tapping are detected using the SD-ROM algorithm, and suppressed using an algorithm approximating the morphology filter. The proposed algorithm was tested using signals collected with a digital stethoscope mockup, and implemented on an ARM920T-based DSP system.