Audio Classification: Environmental sounds classification
Résumé
Recent advancements in the field of machine learning have led to a growing interest in many classification problems, especially involving data in the form of images, video, and audio files. One of the prominent classification problems is to classify sounds and to predict the category of that sound. Some of the applications where such a classification model can be applied in the real world are security systems, classifying music clips to identify the genre of the music, classifying different environmental sounds, speaker detection, and verification. Audio classification is the task of analyzing different audio signals. In this paper, we provide a brief overview of the area of audio classification, describing its system, various modules of feature extraction and modeling, applications, underlying techniques, and some indications of performance. Following this overview, we will discuss some of the strengths and weaknesses of current classification technologies and outline some potential future trends in research, development, and applications. We paid close attention to the inputs, network structures, temporal pooling strategies, and objective functions as these are the fundamental components of many audio classification subtasks. The paper concludes with discussions on future trends and research opportunities in this area.
| Origine | Fichiers produits par l'(les) auteur(s) |
|---|---|
| Licence |