Acoustic Signal Processing and Communications: Topic Modeling of Scopus Abstracts from 2010 till 2024

A determination of major topics in acoustic signal processing and communications using BERTopic, an AI based topic modeling approach, is presented in this article. A total of 3435 abstracts indexed in the SCOPUS database matching a boolean query from year 2010 till mid 2024 were collected using API calls. To compensate for some variability in BERTopic output across runs, 10 repeated runs were performed to find the the topics that remained stable across runs. A total of seven topics were found that occurred in more than one run. Topics “Speech Recognition” and “Underwater Acoustic Communication” were most stable and occurred in all 10 runs. Other topics that appeared in more than 50% of the runs were “Noisy Speech Signal”, “Monitoring System” and “Optical Modulation”. BERTopic based topic modeling offers an insight into major research topics of interest in speech signal processing and communication during the period 2010-2024.