Multichannel environmental sound segmentation

Abstract

This paper proposes a multichannel environmental sound segmentation method. Environmental sound segmentation is an integrated method to achieve sound source localization, sound source separation and classification, simultaneously. When multiple microphones are available, spatial features can be used to improve the localization and separation accuracy of sounds from different directions; however, conventional methods have three drawbacks: (a) Sound source localization and sound source separation methods using spatial features and classification using spectral features trained in the same neural network, may overfit to the relationship between the direction of arrival and the class of a sound, thereby reducing their reliability to deal with novel events. (b) Although permutation invariant training used in autonomous speech recognition could be extended, it is impractical for environmental sounds that include an unlimited number of sound sources. (c) Various features, such as complex values of short time Fourier transform and interchannel phase differences have been used as spatial features, but no study has compared them. This paper proposes a multichannel environmental sound segmentation method comprising two discrete blocks, a sound source localization and separation block and a sound source separation and classification block. By separating the blocks, overfitting to the relationship between the direction of arrival and the class is avoided. Simulation experiments using created datasets including 75-class environmental sounds showed the root mean squared error of the proposed method was lower than that of conventional methods.

Detection technology application based on spectral subtraction and vibro acoustic principle in the measurement of ship reliability level

“…On the basis of all the above experiments, ANSYS, abaqus, and other simulation software are used to carry out modal analysis experiments. The ultimate purpose is to obtain the natural frequency of ship reliability (Sudo et al, 2021). For a continuous initial signal, the calculation obtained by Short-Time Fourier Transform (STFT) is shown in Eq.…”

Sound Classification and Processing of Urban Environments: A Systematic Literature Review

“…Different types of autoencoders allow a multiplicity of applications, such as the one in the work of Sudo et al [36], where a multichannel environmental sound segmentation method comprising the following blocks: Feature Extraction, Sound Source Localization and Separation (SSLS), Sound Source Separation and Classification (SSSC) and reconstruction, was implemented. In this work, the feature extraction is performed using the short-time Fourier transform of the initial input signal, the magnitude spectrograms as spectral features, and the sine and cosine interchannel phase difference as spatial features.…”