Development of Isolated Marathi Words Emotional Speech Database

Abstract

The research work describes the procedure and the development of Isolated Marathi Emotional Speech Database. The database consists of samples, collected from 50 speakers including males and females who simulated the emotions producing by the Marathi utterances which are used in everyday communication and are interpretable in all applied emotions. The speech samples were enhanced by spectral subtraction method and distinguished by the various real life situations. The recorded speech samples were categorized in three basic categories i.e. Happy, Sad and Angry.

Peer Review #1 of "Effect on speech emotion classification of a feature selection approach using a convolutional neural network (v0.1)"

“…Milton, A. et al presented a three-stage traditional SVM classifying different Berlin emotional datasets (Milton et al (2013)). VB Waghmare et al adopted spectral features (MFCCs) as the main feature and classified emotions from the Marathi speech dataset (Waghmare et al (2014)). Demircan, S. et al extracted MFCC features from the Berlin EmoDB database.…”

A Review on Marathi Language Speech Database Development for Automatic Speech Recognition (ASR) System.

“…The paper is organized as follows: Section II describe about Marathi language Section III describe the various speech database developed in Marathi language section IV gives conclusion & future work followed by references. [11].…”