The improved automatic control points computation for the acoustic noise level audits

<p class="Abstract">The acoustic noise level in the interior is one of the quantities specified by a standard and is subject to audits to ensure a comfortable living environment. Currently, the noise level audits are performed manually by a skilled operator, who evaluates the floor plan and uses it to calculate the control points location in which the measurement is performed. The computation is proposed to automate the audit by formulating an optimisation problem for which an algorithm was designed. The algorithm computes the solution that satisfies all constraints specified in the standard, for example, the minimum distance among the control points and fixed obstacles (walls or columns). In the proposed optimisation problem, the fitness function was designed based on the measurement purpose, and two typical use-cases were analysed: (i) long-term stationary noise measurement and (ii) recurring short-term noise measurement. Although the set of control points for both use cases complies with the given standard, it is beneficial to distinguish the location of control points based on the measurement purpose. The number of control points is maximised for the stationary noise and for the immediate coverage area for the short-term noise. The proposed algorithms were tested in a simulation for several floor plans of different complexity.</p>

A noise PSD estimation algorithm using derivative-based high-pass filter in non-stationary noise conditions

The minimum mean-square error (MMSE)-based noise PSD estimators have been used widely for speech enhancement. However, the MMSE noise PSD estimators assume that the noise signal changes at a slower rate than the speech signal— which lacks the ability to track the highly non-stationary noise sources. Moreover, the performance of the MMSE-based noise PSD estimator largely depends upon the accuracy of the a priori SNR estimation in practice. In this paper, we introduce a noise PSD estimation algorithm using a derivative-based high-pass filter in non-stationary noise conditions. The proposed method processes the silent and speech frames of the noisy speech differently to estimate the noise PSD. It is due to the non-stationary noise that can be mixed with silent and speech-dominated frames non-uniformly. We first introduce a spectral-flatness-based adaptive thresholding technique to detect the speech activity of the noisy speech frames. Since the silent frame of the noisy speech is completely filled with noise, the noise periodogram is directly computed from it without applying any filtering. Conversely, a 4th order derivative-based high-pass filter is applied during speech activity of the noisy speech frame to filter out the clean speech components while leaving behind mostly the noise. The noise periodogram is computed from the filtered signal—which counteracts the leaking of clean speech power. The noise PSD estimate is obtained by recursively averaging the previously estimated noise PSD and the current estimate of the noise periodogram. The proposed method is found to be effective in tracking the rapidly changing as well as the slowly varying noise PSD than the competing methods in non-stationary noise conditions for a wide range of signal-to-noise ratio (SNR) levels. Extensive objective and subjective scores on the NOIZEUS corpus demonstrate that the application of the proposed noise PSD with MMSE-based speech enhancement methods produce higher quality and intelligible enhanced speech than the competing methods.

Calculation tool NoBel for sound propagation assessment of noise from gasturbines on the ground

Gas turbine engines of aviation type are the basic element of gas-pumping stations of the pipelines around in the world. They produce harmful noise and the protection zone around the gas-pumping station must provide the safe distance to residence in its vicinity. Due to Ukrainian legal requirements for the stationary (not movable) noise sources the sound pressure levels in octave bands must be used as limits for protection of the population complementarily to equivalent and maximum sound levels at day and night periods. A new calculation tool was designed to assess the effects during sound wave propagation from stationary noise sources like gas-turbine installations, as for homogeneous atmosphere, so as for real meteorological and topographical conditions. The tool provides the possibility to predict the conditions of maximum noise exposure at point of noise control. For homogeneous atmosphere the divergence, air absorption, ground surface reflection and acoustic screens contribute to propagation effects mostly. Their formulation in accordance with current knowledge provides more accurate sound levels assessment with differences ± 2 dBA at the boundaries of protection zone in comparison with existing national legal requirements of noise protection calculations. In real atmosphere temperature and wind (in direction and speed) vary with altitude over the ground surface always. These real conditions provide the dependence of sound speed with altitude, and in consequence the refraction of sound waves. For specific conditions of positive sound refraction the tool predicts the sound levels at the receiver up to 10 dBA higher, at negative sound refraction - on 5-7 less than in homogeneous atmosphere for legal protection distances. Tool NoBel is appropriate to be used for specific airport stationary noise scenarios, like for aircraft engine run-ups after their maintenance or repair.

A neural network based noise suppression method for transient noise control with low-complexity computation

Over the decades, the noise-suppression (NS) methods for speech enhancement (SE) have been widely utilized, including the conventional signal processing methods and the deep neural networks (DNN) methods. Although stationary-noise can be suppressed successfully using conventional or DNN methods, it is significantly challenging while suppressing the non-stationary noise, especially the transient noise. Compared to conventional NS methods, DNN NS methods may work more effectively under non-stationary noises by learning the noises' temporal-frequency characteristics. However, most DNN methods are challenging to be implemented on mobile devices due to their heavy computation complexity. Indeed, even a few low-complexity DNN methods are proposed for real-time purposes, the robustness and the generalization degrade for different types of noise. This paper proposes a single channel DNN-based NS method for transient noise with low computation complexity. The proposed method enhanced the signal-to-noise ratio (SNR) while minimizing the speech's distortion, resulting in a superior improvement of the speech quality over different noise types, including transient noise.