Tai-Hwei Hwang - researchr alias

researchr

You are not signed in
Sign in
Sign up

Viewing Publication 1 - 12 from 12

2014

Rain Removal Using Single Image based on Non-negative Matrix FactorizationPin-Hsian Liu, Chih-Yang Lin, Chia-Hung Yeh, Li-Wei Kang, Kyle Shih-Huang Lo, Tai-Hwei Hwang, Chia-Chen Kuo. intcompsymp 2015: 1137-1146 [doi]

2013

Vehicle trajectory prediction across non-overlapping camera networksChing-Chun Huang, Hung-Nguyen Manh, Tai-Hwei Hwang. iccve 2013: 375-380 [doi]

2005

Combination of autocorrelation-based features and projection measure technique for speaker identificationKuo-Hwei Yuo, Tai-Hwei Hwang, Hsiao-Chuan Wang. taslp, 13(4):565-574, 2005. [doi]

2004

Energy contour enhancement for noisy speech recognitionTai-Hwei Hwang, Sen-Chia Chang. iscslp 2004: 249-252 [doi]

2003

Energy contour extraction for in-car speech recognitionTai-Hwei Hwang. interspeech 2003: 2181-2184 [doi]

2001

Linear interpolation of cepstral variance for noisy speech recognitionTai-Hwei Hwang, Kuo-Hwei Yuo, Hsiao-Chuan Wang. interspeech 2001: 877-880 [doi]

2000

Speech model compensation with direct adaptation of cepstral variance to noisy environmentTai-Hwei Hwang, Kuo-Hwei Yuo, Hsiao-Chuan Wang. interspeech 2000: 366-369 [doi]

A fast algorithm for parallel model combination for noisy speech recognitionTai-Hwei Hwang, Hsiao-Chuan Wang. csl, 14(2):81-100, 2000. [doi]

Combination of temporal trajectory filtering and projection measure for robust speaker identificationKuo-Hwei Yuo, Tai-Hwei Hwang, Hsiao-Chuan Wang. interspeech 2000: 791-794 [doi]

1998

Weighted parallel model combination for noisy speech recognitionTai-Hwei Hwang, Hsiao-Chuan Wang. interspeech 1998: [doi]

1997

Adaptation of time differentiated cepstrum for noisy speech recognitionTai-Hwei Hwang, Lee-Min Lee, Hsiao-Chuan Wang. interspeech 1997: 1075-1078 [doi]

Feature adaptation using deviation vector for robust speech recognition in noisy environmentTai-Hwei Hwang, Lee-Min Lee, Hsiao-Chuan Wang. icassp 1997: 1227-1230 [doi]

Sign in or sign up to see more results.

runs on WebDSL