Please use this identifier to cite or link to this item:
Title: Development of high-performance and large-scale Vietnamese automatic speech recognition systems
Authors: Do, Quoc Truong
Pham, Ngoc Phuong
Tran, Hoang Tung
Luong, Chi Mai
Keywords: ASR
Automatic speech recognition
Vietnamese corpora
Vietnamese Speech recognition
Issue Date: 2018
Series/Report no.: Journal of Computer Science and Cybernetics;Vol.34(04) .- P.335–348
Abstract: Automatic Speech Recognition (ASR) systems convert human speech into corresponding transcription automatically. They have a wide range of application such as controlling robots, call center analytic, voice chatbot. Recent studies on ASR for English have achieved the performance that surpass human ability. The systems were trained on a large amount of training data and performed well under many environments. With regards to Vietnamese, there have been many studies on improving the performance of existing ASR systems, however, many of them are conducted on a small-scaled data, which does not reflect realistic scenarios. Although the corpora used to train the system were carefully design to maintain phonetic balance properties, efforts in collecting them at a large-scale is still limited. Specifically, only a certain accent of Vietnam was evaluated in existing works. In this paper, we first describe our efforts in collecting a large data set that covers all 3 major accents of Vietnam located in the Northern, Center, and Southern regions. Then, we detail our ASR system development procedure utilizing the collected data set and evaluating different model architectures to find the best structure for Vietnamese. In the VLSP 2018 challenge, our system achieved the best performance with 6,5% WER and on our internal test set with more than 10 hours of speech collected real environments, the system also performs well with 11% WER.
ISSN: 1813-9663
Appears in Collections:Tin học và Điều khiển học (Journal of Computer Science and Cybernetics)

Files in This Item:
File Description SizeFormat 
_file_5 MBAdobe PDFView/Open
Your IP:

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.