Yixiang Chen

Yixiang Chen
Are you Yixiang Chen?

Claim your profile, edit publications, add additional information:

Contact Details

Name
Yixiang Chen
Affiliation
Location

Pubs By Year

Pub Categories

 
Computer Science - Computation and Language (4)
 
Computer Science - Learning (4)
 
Computer Science - Sound (2)
 
Computer Science - Neural and Evolutionary Computing (2)
 
Computer Science - Artificial Intelligence (1)

Publications Authored By Yixiang Chen

Recently deep neural networks (DNNs) have been used to learn speaker features. However, the quality of the learned features is not sufficiently good, so a complex back-end model, either neural or probabilistic, has to be used to address the residual uncertainty when applied to speaker verification, just as with raw features. This paper presents a convolutional time-delay deep neural network structure (CT-DNN) for speaker feature learning. Read More

Deep neural models, particularly the LSTM-RNN model, have shown great potential in language identification (LID). However, the phonetic information has been largely overlooked by most of existing neural LID methods, although this information has been used in the conventional phonetic LID systems with a great success. We present a phonetic temporal neural model for LID, which is an LSTM-RNN LID system but accepts phonetic features produced by a phone-discriminative DNN as the input, rather than raw acoustic features. Read More

Pure acoustic neural models, particularly the LSTM-RNN model, have shown great potential in language identification (LID). However, the phonetic information has been largely overlooked by most of existing neural LID models, although this information has been used in the conventional phonetic LID systems with a great success. We present a phone-aware neural LID architecture, which is a deep LSTM-RNN LID system but accepts output from an RNN-based ASR system. Read More

PLDA is a popular normalization approach for the i-vector model, and it has delivered state-of-the-art performance in speaker verification. However, PLDA training requires a large amount of labelled development data, which is highly expensive in most cases. We present a cheap PLDA training approach, which assumes that speakers in the same session can be easily separated, and speakers in different sessions are simply different. Read More