J
Jia Cui
Researcher at Tencent
Publications - 44
Citations - 1095
Jia Cui is an academic researcher from Tencent. The author has contributed to research in topics: Artificial neural network & Language model. The author has an hindex of 17, co-authored 41 publications receiving 861 citations. Previous affiliations of Jia Cui include IBM & Peking University.
Papers
More filters
Proceedings ArticleDOI
Efficient Knowledge Distillation from an Ensemble of Teachers.
TL;DR: It is shown that with knowledge distillation, information from multiple acoustic models like very deep VGG networks and Long Short-Term Memory models can be used to train standard convolutional neural network (CNN) acoustic models for a variety of systems requiring a quick turnaround.
Journal ArticleDOI
Post-Transcriptional Regulation of Anti-Apoptotic BCL2 Family Members
Jia Cui,William J. Placzek +1 more
TL;DR: The RNA binding proteins (RBPs) and microRNAs (miRNAs) that mediate post-transcriptional regulation of the anti-apoptotic BCL2 family members are discussed and their roles and impact on alternative splicing, mRNA turnover, and mRNA subcellular localization are described.
Proceedings ArticleDOI
Multilingual representations for low resource speech recognition and keyword search
Jia Cui,Brian Kingsbury,Bhuvana Ramabhadran,Abhinav Sethy,Kartik Audhkhasi,Xiaodong Cui,Ellen Kislal,Lidia Mangu,Markus Nussbaum-Thom,Michael Picheny,Zoltán Tüske,Pavel Golik,Ralf Schlüter,Hermann Ney,Mark J. F. Gales,Kate Knill,Anton Ragni,Haipeng Wang,P.C. Woodland +18 more
TL;DR: This paper examines the impact of multilingual acoustic representations on Automatic Speech Recognition (ASR) and keyword search (KWS) for low resource languages in the context of the OpenKWS15 evaluation of the IARPA Babel program and shows that these multilingual representations significantly improve ASR and KWS performance.
Proceedings ArticleDOI
Improving Attention Based Sequence-to-Sequence Models for End-to-End English Conversational Speech Recognition.
TL;DR: This work proposes to use an input-feeding architecture which feeds not only the previous context vector but also the previous decoder hidden state information as inputs to the decoder, based on a better hypothesis generation scheme for sequential minimum Bayes risk (MBR) training of sequence-to-sequence models.
Proceedings ArticleDOI
System combination and score normalization for spoken term detection
Jonathan Mamou,Jia Cui,Xiaodong Cui,Mark J. F. Gales,Brian Kingsbury,Kate Knill,Lidia Mangu,David Nolden,Michael Picheny,Bhuvana Ramabhadran,Ralf Schlüter,Abhinav Sethy,Philip C. Woodland +12 more
TL;DR: This paper investigates the problem of extending data fusion methodologies from Information Retrieval for Spoken Term Detection on low-resource languages in the framework of the IARPA Babel program, and describes a number of alternative methods improving keyword search performance.