Speaker Independent Isolated Tamil Words Recognition System using Different Classifiers
Journal Title: International Journal of Computer Science & Engineering Technology - Year 2015, Vol 6, Issue 6
Abstract
The paper is to build a speaker independent isolated speech recognition system for Tamil Language. The system consists of four steps: first step involves recording of Tamil spoken words using high quality studio microphone and stored in the form of wav files. The next step is preprocessing, to filter out noise present in speech signal using pre-emphasis filter. The filtered speech signal features are extracted using Mel Frequency Cepstral Coefficient (MFCC) technique. The extracted features from spoken Tamil words are fed as input to different classifiers such as Probabilistic neural network, Hidden markov model and Support Vector Machine. Each Classifier recognition performance is analyzed, and found that hidden markov model achieves higher recognition rate than other classifiers.
Authors and Affiliations
P. Iswarya , Dr. V. Radha
Keywords
Related Articles
- EP ID EP163776
- DOI -
- Views 120
- Downloads 0
Impact of Protocol Switching Strategy in Delay Tolerant Networks
Delay-Tolerant Networks (DTNs) are sparse dynamic wireless networks in which a complete end-to-end path from the source to the destination does not exist. In this context, conventional mobile adhoc routing schemes would...
COMPARATIVE ANALYSIS OF DIFFERENT ROUTING PROTOCOLS IN DELAY TOLERANT NETWORKS
The modern wireless networks have provided a wide range of applications making it possible to successfully interconnect devices and systems which have the wireless connection facility, such as mobile phones, tablets, lap...
A Comparative Analysis of Energy Preservation Performance Metric for ERAODV, RAODV, AODV and DSDV Routing Protocols in MANET
In Mobile Ad-Hoc Networks (MANETs), mobile devices are operational with power. In order to utilize these energy operational devices resourcefully for transmission of data packets, many energy aware routing strategies are...
Enhancing Efficiency and Accuracy of Imbalanced Datasets Using Fuzzy Neural Network
In Data Mining the class Imbalance classification problem is considered to be one of the emergent challenges. This problem occurs when the number of examples that represents one of the classes of the dataset is much lowe...
Benefits of Learning Management System (LMS) in Indian Education
In today’s scenario, academic institutions are striving to find the right combination of students, faculties, protocols and systems to manage their learning programs. Learning management systems (LMS) are becoming an int...