Endpoint detection enhancement for speaker dependent recognition
The automatic speech recognition (ASR) field has become one of the leading speech technology areas today. Various methods have been introduced to develop an efficient ASR system. The Neural Network (NN) approach is one of the more popular methods that is widely used in this field. Another Multilayer...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Universiti Kebangsaan Malaysia Press
2009
|
| Online Access: | http://psasir.upm.edu.my/id/eprint/14497/ http://psasir.upm.edu.my/id/eprint/14497/1/Endpoint%20detection%20enhancement%20for%20speaker%20dependent%20recognition.pdf |
| _version_ | 1848842410287693824 |
|---|---|
| author | Mohamad Hussin, Ummu Salmah Shamsuddin, Siti Maryam Mahmud, Ramlan |
| author_facet | Mohamad Hussin, Ummu Salmah Shamsuddin, Siti Maryam Mahmud, Ramlan |
| author_sort | Mohamad Hussin, Ummu Salmah |
| building | UPM Institutional Repository |
| collection | Online Access |
| description | The automatic speech recognition (ASR) field has become one of the leading speech technology areas today. Various methods have been introduced to develop an efficient ASR system. The Neural Network (NN) approach is one of the more popular methods that is widely used in this field. Another Multilayer perceptron (MLP) model which is popularly used in the ASR field is the NN model. However, the current problems faced by MLP and most NN models in the ASR field is the long duration of training. Furthermore, the robustness of the isolated digit recognition is not trivial because it has been widely used in many applications. This study focuses on improving the training time and robustness of the MLP neural network for the Malay isolated digit recognition system by proposing variance endpoint detection to accelerate the convergence time of the NN and to produce the highest recognition accuracy. The proposed endpoint method have shown very promising results over experiments carried out. The overall performance for the Malay data set is 99.83% with a convergence time of 82 seconds. |
| first_indexed | 2025-11-15T07:58:41Z |
| format | Article |
| id | upm-14497 |
| institution | Universiti Putra Malaysia |
| institution_category | Local University |
| language | English |
| last_indexed | 2025-11-15T07:58:41Z |
| publishDate | 2009 |
| publisher | Universiti Kebangsaan Malaysia Press |
| recordtype | eprints |
| repository_type | Digital Repository |
| spelling | upm-144972016-02-07T13:30:47Z http://psasir.upm.edu.my/id/eprint/14497/ Endpoint detection enhancement for speaker dependent recognition Mohamad Hussin, Ummu Salmah Shamsuddin, Siti Maryam Mahmud, Ramlan The automatic speech recognition (ASR) field has become one of the leading speech technology areas today. Various methods have been introduced to develop an efficient ASR system. The Neural Network (NN) approach is one of the more popular methods that is widely used in this field. Another Multilayer perceptron (MLP) model which is popularly used in the ASR field is the NN model. However, the current problems faced by MLP and most NN models in the ASR field is the long duration of training. Furthermore, the robustness of the isolated digit recognition is not trivial because it has been widely used in many applications. This study focuses on improving the training time and robustness of the MLP neural network for the Malay isolated digit recognition system by proposing variance endpoint detection to accelerate the convergence time of the NN and to produce the highest recognition accuracy. The proposed endpoint method have shown very promising results over experiments carried out. The overall performance for the Malay data set is 99.83% with a convergence time of 82 seconds. Universiti Kebangsaan Malaysia Press 2009 Article PeerReviewed application/pdf en http://psasir.upm.edu.my/id/eprint/14497/1/Endpoint%20detection%20enhancement%20for%20speaker%20dependent%20recognition.pdf Mohamad Hussin, Ummu Salmah and Shamsuddin, Siti Maryam and Mahmud, Ramlan (2009) Endpoint detection enhancement for speaker dependent recognition. Asia-Pacific Journal of Information Technology and Multimedia, 7. pp. 17-29. ISSN 2289-2192 http://ejournals.ukm.my/apjitm/article/view/1290/1160 |
| spellingShingle | Mohamad Hussin, Ummu Salmah Shamsuddin, Siti Maryam Mahmud, Ramlan Endpoint detection enhancement for speaker dependent recognition |
| title | Endpoint detection enhancement for speaker dependent recognition |
| title_full | Endpoint detection enhancement for speaker dependent recognition |
| title_fullStr | Endpoint detection enhancement for speaker dependent recognition |
| title_full_unstemmed | Endpoint detection enhancement for speaker dependent recognition |
| title_short | Endpoint detection enhancement for speaker dependent recognition |
| title_sort | endpoint detection enhancement for speaker dependent recognition |
| url | http://psasir.upm.edu.my/id/eprint/14497/ http://psasir.upm.edu.my/id/eprint/14497/ http://psasir.upm.edu.my/id/eprint/14497/1/Endpoint%20detection%20enhancement%20for%20speaker%20dependent%20recognition.pdf |