Acoustic model, language model, MFCC, HMM, Word error rate, Deep Neural Networks, Recurrent Neural Networks, Convolutional Neural Networks
Abstract
Language plays a significant role in human life as it allows us to communicate, express ourselves and understand others. It is the primary means of communication. Many research activities are conducted on automatic speech recognition. The Major drawback of the ASR systems is their Efficiency. The overall performance of the ASR system Depends mostly on the Acoustic model and also affected by the environment. For ASR systems Mostly we use Deep learning techniques like Recurrent neural networks used for speech recognition, voice recognition, time series prediction and natural language processing and Convolutional Neural Network which uses different modules for speech emotion recognition and classifiers are used to Differentiate emotions such as happiness, surprise, Anger, neural state, sadness etc.
Article Details
Unique Paper ID: 159931
Publication Volume & Issue: Volume 9, Issue 12
Page(s): 863 - 866
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024