Kannada Speech Emotion Recognition Using Ensembling Techniques
  • Author(s): Smrithi Baliga ; Sapna H M ; Shreyas N ; Yogesh Gowda V ; Dr Chandrashekar M Patil; Prof. Audre Arlene
  • Paper ID: 1704436
  • Page: 250-255
  • Published Date: 11-05-2023
  • Published In: Iconic Research And Engineering Journals
  • Publisher: IRE Journals
  • e-ISSN: 2456-8880
  • Volume/Issue: Volume 6 Issue 11 May-2023
Abstract

This study explores the development of a speech emotion recognition system for the Kannada language, using a dataset of audio recordings labeled with six emotion categories: happiness, sadness, anger, fear, and neutral. We used a combination of acoustic features and machine learning algorithms, including Mel-frequency cepstral coefficients (MFCCs), to classify emotions in the audio recordings. Our results show that the proposed system achieves an average accuracy of 75% on the Kannada emotion dataset, outperforming existing baseline models. These findings suggest that Kannada speech emotion recognition can be achieved with high accuracy using a combination of acoustic features and machine learning algorithms like RNN, CNN and DBN, paving the way for further research in this area.

Keywords

Speech Emotion Recognition, Mel-Frequency Cepstral Coefficients, Recurrent Neural Network, Deep Belief Network

Citations

IRE Journals:
Smrithi Baliga , Sapna H M , Shreyas N , Yogesh Gowda V , Dr Chandrashekar M Patil; Prof. Audre Arlene "Kannada Speech Emotion Recognition Using Ensembling Techniques" Iconic Research And Engineering Journals Volume 6 Issue 11 2023 Page 250-255

IEEE:
Smrithi Baliga , Sapna H M , Shreyas N , Yogesh Gowda V , Dr Chandrashekar M Patil; Prof. Audre Arlene "Kannada Speech Emotion Recognition Using Ensembling Techniques" Iconic Research And Engineering Journals, 6(11)