Voice and speech recognition in Tamil language

摘要

In our project, our intention is to create a voice and speech recognition system in smart phones that recognizes voice and captures the speech data in Tamil and stores and converts the captured speech as text in Tamil language itself. This can be used in voice dialing, sending SMS by saying out the message and the captured message is sent to the recipient in Tamil. There has not been much consideration for Tamil language to be used in voice and speech recognition in smart phones. For native users Tamil voice recognitions and speech would provide more flexibility in smart phone experience. Also people who have only been used to their native language Tamil, would feel easier to use the speech recognition system in their smart phones if provided in Tamil. There will be no more difficulty in usage of phones for local users and there is no need for any learning to use the smart phone. Automatic Speech Recognition (ASR) system have achieved a great success in many applications. Among them, Template Matching techniques like Dynamic Time Warping (DTW), Statistical Pattern Matching techniques such as Hidden Markov Model (HMM) and Gaussian Mixture Models (GMM), Machine Learning techniques such as Neural Networks (NN), Support Vector Machine (SVM), and Decision Trees (DT) are most popular. For this system, highest word recognition accuracy is achieved with HMM technique. It offered 100% accuracy during training process and approximately 98% for testing process.