Punjabi Speech Recognition of Isolated Words Using Compound EEMD & Neural Network
AnchalKatyal1, Amanpreet Kaur2, Jasmeen Gill3
1AnchalKatyal, M.Tech (CSE) RIMT-IET Mandi Gobindgarh, India.
2Amanpreet Kaur, Assistant Professor BBSBEC Fatehgarh Sahib, India.
3Jasmeen Gill, Assistant Professor RIMT-IET Mandi Gobindgarh, India.
Manuscript received on March 01, 2014. | Revised Manuscript received on March 04, 2014. | Manuscript published on March 05, 2014. | PP: 150-154 | Volume-4 Issue-1, March 2014. | Retrieval Number: A2063034114/2014©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Automatic Speech recognition and conversion of speech to text is a work which has proved its importance for decades. A lot of work has already been done in this contrast. This paper focuses on the Punjabi speech and the conversion of speech to text using advanced system voice recognition pattern. This paper also focuses on the optimization of the EEMD process by combining EEMD process with the Neural Network. Neural Network has been found to be friendly in the contrast of compounding different algorithms to it and it produces significant results. This paper also focuses on the future works to be considered in the same field.
Keywords: ASR, EEMD, Neural Network, Acoustical Models, Neural Identifier, Data Acquisition.