Different Speech Recognition EMG based Approaches for SSI |
Author(s): |
| Geeta N. Sonawane , SGDCOE, Jalgaon; Mr. C. S. Pati, SGDCOE, Jalgaon; Mrs. A.N.Shewale, SGDCOE, Jalgaon; Mr. R. R. Karhe, SGDCOE, Jalgaon; Gunjan G. Gujarathi, SGDCOE, Jalgaon |
Keywords: |
| Articulatory Muscles, Electromyography-Based Speech Recognition, Modalities, SEMG, Silent Speech Interface |
Abstract |
|
This paper reflects the different sampled data used in speech recognition for SSI. The EMG is an electrical potential generated by movements associated with speech sound production in glottal region, jaw, tongue, soft palate, lips and other areas. This paper provides up-to date review of current approaches based on speech recognition EMG based speech recognition. The techniques are helpful to enable SSI, when EMG signals are recorded for those people who only articulate speech without producing any sound. The SSI have mostly depends on sampled data for facial and visual modalities to reduce the negative effect in speaker variation and recognition accuracy. The outline of discussion that it will need to improve in large training dataset, used of different languages based model, removal of noise artifacts. |
Other Details |
|
Paper ID: IJSRDV4I50840 Published in: Volume : 4, Issue : 5 Publication Date: 01/08/2016 Page(s): 1546-1550 |
Article Preview |
|
|
|
|
