MA, Duo
POSITION/TITLE
SRIBD Research Engineer
RESEARCH FIELD
Speech Recognition, Speech Enhancement, Speech Processing, Speech-Text Multimodal Pretraining
maduo@sribd.cn
PERSONAL WEBSITE
https://shanguanma.github.io/
EDUCATION BACKGROUND
M.Sc. Northwest Minzu University
B.Sc. Huanghe S & T University
BIOGRAPHY
Duo Ma is currently a research engineer at the Human Language Technology Laboratory of the Shenzhen Research Institute of Big Data. Prior to that, he was a speech research engineer at the Technology Platform Center of Purdue Technology, where he was mainly engaged in the research and landing optimization of speech recognition, voice wake-up and voice command for food service robots in high-noise and complex scenarios. Master's Degree is from National Languages Information Technology (NLIT), Northwest Minzu University, China. During his master's degree, he was engaged in speech recognition and speech enhancement research, and during his master's degree, he went to Temasek Labs, Nanyang Technological University, Singapore, for two years as an academic exchange visitor. After graduation, he worked as a research engineer in HLT Laboratory, National University of Singapore, engaging in speech recognition, speech enhancement and other research work. Duo Ma is the winner of IEEE SLT 2021 Alpha-mini Speech Challenge(ASC)，Keyword Spotting(KWS) Track.
ACADEMIC PUBLICATIONS
Duo Ma, Nana Hou, VanTung Pham, Haihua Xu, and EngSiong Chng. "Multitask-based joint learning approach to robust asr for radio communication speech." 2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC). IEEE, 2021.
Chen. Chen, Nana Hou, Duo Ma and EngSiong Chng, "Time Domain Speech Enhancement With Attentive Multi-scale Approach,"2021 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC), 2021, pp. 679-683.
Duo Ma, Guanyu Li, Haihua Xu, and Eng Siong Chng. 2019. “Improving Code-Switching Speech Recognition with Data Augmentation and System Combination.” In 2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC).
