Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Haihua Xu

Leveraging Text Data Using Hybrid Transformer-LSTM Based End-to-End ASR in Transfer Learning


May 28, 2020
Zhiping Zeng, Van Tung Pham, Haihua Xu, Yerbolat Khassanov, Eng Siong Chng, Chongjia Ni, Bin Ma


  Access Paper or Ask Questions

Approaches to Improving Recognition of Underrepresented Named Entities in Hybrid ASR Systems


May 18, 2020
Tingzhi Mao, Yerbolat Khassanov, Van Tung Pham, Haihua Xu, Hao Huang, Eng Siong Chng


  Access Paper or Ask Questions

Spatial-Scale Aligned Network for Fine-Grained Recognition


Jan 05, 2020
Lizhao Gao, Haihua Xu, Chong Sun, Junling Liu, Yu-Wing Tai


  Access Paper or Ask Questions

Independent language modeling architecture for end-to-end ASR


Nov 25, 2019
Van Tung Pham, Haihua Xu, Yerbolat Khassanov, Zhiping Zeng, Eng Siong Chng, Chongjia Ni, Bin Ma, Haizhou Li


  Access Paper or Ask Questions

Constrained Output Embeddings for End-to-End Code-Switching Speech Recognition with Only Monolingual Data


Apr 08, 2019
Yerbolat Khassanov, Haihua Xu, Van Tung Pham, Zhiping Zeng, Eng Siong Chng, Chongjia Ni, Bin Ma

* 5 pages, 3 figures, submitted to INTERSPEECH 2019 

  Access Paper or Ask Questions

Enriching Rare Word Representations in Neural Language Models by Embedding Matrix Augmentation


Apr 08, 2019
Yerbolat Khassanov, Zhiping Zeng, Van Tung Pham, Haihua Xu, Eng Siong Chng

* 5 pages, 2 figures, submitted to INTERSPEECH 2019 

  Access Paper or Ask Questions

On the End-to-End Solution to Mandarin-English Code-switching Speech Recognition


Nov 01, 2018
Zhiping Zeng, Yerbolat Khassanov, Van Tung Pham, Haihua Xu, Eng Siong Chng, Haizhou Li

* Submitted to ICASSP 2019 

  Access Paper or Ask Questions

Study of Semi-supervised Approaches to Improving English-Mandarin Code-Switching Speech Recognition


Jun 16, 2018
Pengcheng Guo, Haihua Xu, Lei Xie, Eng Siong Chng

* 5pages, 3 figures, INTERSPEECH 2018 

  Access Paper or Ask Questions