Loading...
Please wait, while we are loading the content...
Hindi Phoneme Recognition Using Time Delay Neural Network
| Content Provider | Semantic Scholar |
|---|---|
| Author | Sarma, A. V. R. S. Strube, Hans Werner Agrawal, Shyam Sundar |
| Copyright Year | 2007 |
| Abstract | INTRODUCTION Recent work on neural networks raises the possibility of new approaches to the speech recognition problem [1]-[4]. Currently, most speech recognition systems are based on Hidden Markov Models (HMMs), a statistical framework that supports both acoustic and temporal modelling. Despite their state-of-the-art performance, HMMs make a number of suboptimal modelling assumptions that limit their potential effectiveness. Neural networks avoid many of these assumptions, while they can also learn complex functions, generalize effectively, tolerate noise, and support parallelism. Several speech researchers implemented neural networks for speech recognition and the results were compared with the other available pattern classifiers. The Time Delay Neural Network (TDNN) has been successfully implemented by several speech researchers for the recognition of English and Japanese phonemes [5]-[7]. It has the ability to 1.) represent temporal relationships between acoustic events, while at the same time 2.) provide for the invariance under translation in time. Recently Hagelstange and Knohl [8] of University of Bochum, Germany have implemented TDNN techniques for the recognition of German phonemes and demonstrated superior performance. So far these techniques have not been applied to Indian languages. The authors have implemented TDNN (based on the Bochum system) for the recognition of unvoiced unaspirated and voiced unaspirated stop consonants of Hindi speech, and the results are presented in this paper. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | http://www.physik3.gwdg.de/~strube/papers/daga97sarma.pdf |
| Alternate Webpage(s) | http://www.physik3.gwdg.de/Personen/Gaeste/Sarma/daga97-19951001.ps |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
