Composite feature set for mood recognition in dialectal Assamese speech
Authors:
- 1 February 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 691-695
Abstract
Speech is a rich source of information. The speech samples can not only retain what is being spoken but also the emotional state of the speaker. In this paper, the dynamics of the prosodic features and the spectral features have been used to encode the mood content of speakers of Assamese language with dialectal components. A composite feature set has been created by fusing the spectral and the prosodic features. The performance of the system has been evaluated using two classifiers namely Recurrent Neural Network (RNN) and Feed Forward Time Delay Neural Network (FFTDNN). A comparative analysis has been made on their computational speed and recognition rates. The performance of the proposed mood verification system has also been evaluated by varying the background noise conditions.Keywords
This publication has 7 references indexed in Scilit:
- Phoneme-Based Speech Segmentation using Hybrid Soft Computing Framework2014
- A Study of Prosodic Features of Emotional Speech2012
- Role of neural network models for developing speech systemsSādhanā, 2011
- Survey on speech emotion recognition: Features, classification schemes, and databasesPattern Recognition, 2011
- Characterization of emotions using the dynamics of prosodic features2010
- Extraction and representation of prosodic features for language and speaker recognitionSpeech Communication, 2008
- Introduction to Digital Speech ProcessingFoundations and Trends® in Signal Processing, 2007