C
Chia-Ping Chen
Researcher at National Sun Yat-sen University
Publications - 54
Citations - 644
Chia-Ping Chen is an academic researcher from National Sun Yat-sen University. The author has contributed to research in topics: Feature extraction & Artificial neural network. The author has an hindex of 10, co-authored 54 publications receiving 595 citations. Previous affiliations of Chia-Ping Chen include University of Washington.
Papers
More filters
Journal ArticleDOI
MVA Processing of Speech Features
Chia-Ping Chen,Jeff A. Bilmes +1 more
TL;DR: In this paper, mean subtraction, variance normalization, and auto-regression moving-average (ARMA) filtering are applied in the cepstral domain to reduce the distortion of mel-frequency cepSTral coefficients.
Proceedings Article
Low-resource noise-robust feature post-processing on Aurora 2.0.
TL;DR: This work presents a highly effective and extremely simple noiserobust front end based on novel post-processing of standard MFC C features that performs remarkably well on the Aurora 2.0 noisydigits database without requiring any increase in model complexity.
Proceedings Article
Frontend post-processing and backend model enhancement on the Aurora 2.0/3.0 databases.
TL;DR: A highly effective and extremely simple noiserobust front end based on novel post-processing of standard MFCC features on the Aurora databases performs remarkably well on both the Aurora 2.0 and Aurora 3.0 databases without requiring any increase in model complexity.
Proceedings ArticleDOI
Effective Attention Mechanism in Dynamic Models for Speech Emotion Recognition
Po-Wei Hsiao,Chia-Ping Chen +1 more
TL;DR: This work proposes to integrate the attention mechanism into deep recurrent neural network models for speech emotion recognition, based on the intuition that it is beneficial to emphasize the expressive part of the speech signal for emotion recognition.
Proceedings ArticleDOI
Speech feature smoothing for robust ASR
TL;DR: It appears that the effectiveness of normalization and smoothing depends on the domain in which it is applied, being most fruitfully applied just before being scored by a probabilistic model.