A Systemic Review on Automatic Acoustic Scene Classification

Speech research is marked as a most challenging areas among the several challenging research areas. The present literature is analyzed and projected to aid future investigations of the global speech research community. The problems concerned with the acoustic scene corpora, front end algorithms yielding to efficient scene representation, back end engines which outright chore of recognition, are projected in this study. Many works on acoustic scene recognition are analyzed from the perspective of machine learning and deep learning, time length, conviction of development and accessibility. The powerful techniques tending to rich features and robust scene recognition are enlightened in this research review.