Developing a speech corpus from web news for Myanmar (Burmese) language
Citations Over TimeTop 10% of 2017 papers
Abstract
Speech corpus is important for statistical model based automatic speech recognition and it reflects the performance of a speech recognizer. Although most of the speech corpora for resource-riched languages such as English are widely available and it can be used easily, there is no Myanmar speech corpus which is freely available for automatic speech recognition (ASR) research since Myanmar is a low resource language. This paper presents the design and development of Myanmar speech corpus for the news domain to be applied to convolutional neural network (CNN)-based Myanmar continuous speech recognition research. The speech corpus consists of 20 hours read speech data collected from online web news and there are 178 speakers (126 females and 52 males). Our speech corpus is evaluated on two test sets: TestSet1 (web data) and TestSet2 (news recording with 10 natives). Using CNN-based model, word error rate (WER) achieves 24.73% on TestSet1 and 22.95% on TestSet2.
Related Papers
- → Sphinx4 for Indonesian continuous speech recognition system(2017)10 cited
- Preparation of Bangla Speech Corpus from Publicly Available Audio & Text(2020)
- → Corpus Design for Malay Corpus-based Speech Synthesis System(2009)13 cited
- → Recent Progress of Mandrain Spontaneous Speech Recognition on Mandrain Conversation Dialogue Corpus(2019)1 cited
- → CLASSIFICATIONAL PARADIGM OF A TEXT CORPUS BY ITS DESIGN, STRUCTURE AND USE, AS WELL AS BY THE FIXATION AND INDEXATION METHODS OF ITS TEXT DATA(2020)