AI in Audio Analysis: Spectrogram-Based Recognition

Dear Colleagues,

The general field of machine hearing [1] involves algorithms capable of interpreting and extracting meaning from auditory information, similar to how humans recognize sounds, voices, environments, and activities from sound. This research field encompasses sound event detection and classification, auditory scene analysis, forensic audio analysis, medical diagnosis from sound, and more. It includes speech-related tasks such as language and dialect identification, speaker identification, and emotion recognition. Automatic and machine learning approaches, collectively classified as AI, have achieved remarkable performance gains in recent years by representing one-dimensional single-channel audio as a two-dimensional spectrogram [2], whether linear, logarithmic, mel-scaled, constant-Q, stacked filterbanks, or encoded in other ways. Adding a dimension has allowed researchers to unlock the considerable power of image-processing AI techniques and is now commonly used in systems such as the audio spectrum transformer (AST) [3].

In this Special Issue, we explore and extend the field of spectrogram-based recognition. High-quality original research papers are sought in areas including (but not limited to) the following:

• Applications of spectrogram-based audio classification;
• Audio spectrogram-based regression;
• Spectrogram-like representations for deep learning;
• Audio feature transformation of spectrograms;
• Efficient spectrogram-based recognition;
• Spectrograms in speech analysis, enhancement, and coding;
• Anomaly detection from spectral representations;
• Speech and medical applications of audio spectrogram analysis.

[1] Richard F. Lyon, “Machine hearing: An emerging field”, IEEE signal processing magazine 27 (5), 131-139.

[2] H. Zhang, I. McLoughlin and Y. Song, "Robust sound event recognition using convolutional neural networks," 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2015, pp. 559-563, doi: 10.1109/ICASSP.2015.7178031.

[3] Gong, Yuan, Yu-An Chung, and James Glass. "Ast: Audio spectrogram transformer." arXiv preprint arXiv:2104.01778 (2021).

Prof. Dr. Ian McLoughlin
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• sound event detection and classification
• sound scene detection
• acoustic scene analysis
• machine hearing
• spectrograms
• spectral estimation
• audio spectrum transformer
• acoustic feature maps