Abstract
Some people cannot produce sound although their facial muscles work properly due to having problem in their vocal cords. Therefore, recognition of alphabets as well as sentences uttered by these voiceless people is a complex task. This paper proposes a novel method to solve this problem using non-invasive surface Electromyogram (sEMG). Firstly, eleven Bangla vowels are pronounced and sEMG signals are recorded at the same time. Different features are extracted and mRMR feature selection algorithm is then applied to select prominent feature subset from the large feature vector. After that, these prominent features subset is applied in the Artificial Neural Network for vowel classification. This novel Bangla vowel classification method can offer a significant contribution in voice synthesis as well as in speech communication. The result of this experiment shows an overall accuracy of 82.3 % with fewer features compared to other studies in different languages.
Highlights
Language is a powerful tool for self-expression and communication among humans
Hidden Markov Models (HMM) based classifier and a nearest-neighbor classifier based on Dynamic Time Warping (DTW) are studied in (Mondal et al 2009)
The time domain, frequency domain, time–frequency domain features were employed for the classification using Artificial neural networks (ANN)
Summary
Language is a powerful tool for self-expression and communication among humans. Human language is unique compared to other living creatures of the universe in terms of grammatical and semantic categories (Hockett 1960; Deacon 1997) and the property of recursion (Hauser et al 2002). During normal speech, vocal cords in the larynx vibrate and sound is produced just like a musical instrument. Bengali or Bangla comes from Indo-Aryan language and became the seventh mostly spoken language (Summary by Language Size 2013). It is spoken by 193 million people around the world (Summary by Language Size 2013). Some researches have been done in Bangla alphabet recognition, the most of them are in written or acoustic signal; such as, direction code feature based and hidden Markov model based. Hidden Markov Models (HMM) based recognition scheme are used to detect online Bangla handwritten basic characters (Bhattacharya et al 2007; Parui et al 2008). Fuzzy logic is used to classify the Bangla vowels (Kamal et al 2008)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
