Sinhala Speech to Sinhala Unicode Text Conversion for Disaster Relief Facilitation in Sri Lanka

During a stressful situation many people tend to use their own native language. Most of the calls for assistance during a disaster in Sri Lanka is in Sinhala. It takes some time to process this information and determine the type of aid to be sent to the callers. Speech Recognition, also known as automatic speech recognition or computer speech recognition is a process of decoding acoustic speech signal captured by microphone or telephone, to a set of words. Using speech recognition computers are able to understand the voice command of a user and perform the required task. Speech recognition technology is a fast growing engineering technology due to its many uses in various areas. The objective of this study is to design and develop a mechanism that will convert Sinhala speech into Sinhala Unicode text document. This study implements a user-friendly and accurate speech recognition system for the Sinhala language. Hidden Markov Model was used for recognizing the words. The implemented software was able to achieve an accuracy of 65% in a noiseless environment and 54^% in a noisy environment.