Currently, there are three mainstream audio representations, namely channel-based audio, object-based audio, and scene-based audio. The features of content expression differ among these audio representations, the details of which have been specified in the International Telecommunication Union: Radiocommunication Sector (ITU-R) Recommendations. The effective use of these audio representations in accordance with what is to be expressed in the content requires a deep understanding of the technical specifications and capabilities of the audio representations. This review first traces the evolution of loudspeaker layouts developed in recent years, i.e., a history of multichannelization, which is indispensable for the understanding of audio representations. Then, the position of each audio representation among various audio-related standards is described and the method of adopting and implementing each audio representation in other audio-related standards is reviewed using the Moving Picture Experts Group (MPEG) standards as examples.

An acoustic metasurface (AMS) can provide high broadband sound absorption realized by planar periodic array assembly of small Helmholtz resonators tuned at different resonant frequencies. The formulation for the sound absorption coefficient of the AMS developed in a previous study provides a uniform and nearly perfect sound absorption in the one octave band. Numerical case studies suggest that the area ratio of the unit cell assembly for the broadband perfect sound absorption of the AMS can be formulated as a power function of frequency ratio to the center frequency and that the formulation will be identical for any center frequency.

Two finite element models have recently been utilized to predict the sound absorption coefficient of periodically arranged slit resonators. One is based on the linearized Navier–Stokes equation, entropy conservation law, and mass conservation law. The other is based on the Helmholtz equation with the visco-thermal boundary layer boundary condition. This paper clarifies a condition under which the latter model gives a reasonable approximation. A transition frequency is derived from Johnson–Allard's effective density of porous media as a criterion. Calculation results demonstrate that the resonator's resonance frequency must be higher than the transition frequency to obtain a reasonable prediction using the boundary condition model.

Reducing the burden of data collection is crucial for advancing speech recognition research. Hence, this research focuses on exploring methods to enhance machine learning from limited data by augmenting the training data based on three-dimensional measurements in the field of Japanese silent speech recognition. We compared the connectionist temporal classification losses during training and the recognition performance with and without key data augmentation techniques to evaluate the effectiveness of the proposed method utilizing the direct linear transformation method. In this case, the deep neural network was trained successfully, resulting in a reduced phoneme error rate.

Recently, physics-informed neural networks (PINNs) have garnered attention for use as a numerical simulation method for inverse analysis, such as property identification. However, studies on PINNs for conducting acoustic analysis are scarce. Thus, this study developed PINNs that performed acoustic analysis of the vocal tract and synthesized voiced sounds. In addition, PINNs were used to identify glottal source waveforms. Consequently, PINNs were demonstrated to be a promising solution for the inverse problem related to speech production.