Recognition and Classification of Accompanying Audios of Kathak Dance

EasyChair Preprint 6197

Abstract

Here a novel method is been implemented to recognize and classify the accompanying audios used in Kathak Dance. To achieve this Ladi dataset have been used. This audio dataset has been analysed and feature extraction is done. Thus audio and speech processing plays an important part here. Using this dataset beat detection is carried out to detect onset of the beats and then beat tracking is done. Then recognition and classification of Ladi audios is done based on rhythms and mfccs into 7 classes namely L2, L3, L4, L5, L6, L7 and L8 using machine learning based approach.

Keyphrases: Audio and Speech Processing, Audio recognition, Kathak dance, audio analysis, audio classification, feature extraction, k-nearest neighbors, machine learning

@booklet{EasyChair:6197,
  author    = {Anirban Saha and Indranil Sarkar},
  title     = {Recognition and Classification of Accompanying Audios of Kathak Dance},
  howpublished = {EasyChair Preprint 6197},
  year      = {EasyChair, 2021}}