Improving Speech-to-Text recognition for Malaysian english accents using accent identification

• Main Author: Len, Shu Yuan
• Format: Final Year Project / Dissertation / Thesis
• Published: 2022
• Subjects: Q Science (General); T Technology (General)
• Online Access: http://eprints.utar.edu.my/4657/1/fyp_CS_2022_LSY.pdf; http://eprints.utar.edu.my/4657/
• Institution: Universiti Tunku Abdul Rahman

Automatic Speech Recognition (ASR) is the technology that helps user to use their voice as a form of input and it is used in many areas such as mobile devices, embedded systems, and other industrial areas. However, performance and accuracy of the speech recognition system is heavily influenced by the non-native accents, for example, Malaysian English. In this project, the Accent Identification (AID) techniques will be implemented to improve the performance of the ASR systems in recognizing Malaysian English accents. Kaldi toolkits is used in developing proposed ASR models (GMM-HMM and DNN-HMM). CNN based AID is implemented using Python language. The datasets used in this project are from Mini Librispeech, Speech Accent Achieve and other Malaysian English speakers. Then, CNN based AID will be developed and the results is investigated and compared. The Word Error rate is selected as the evaluation metric to compare the recognition performance and accuracy.