13 citations · 75 across the 46 of their papers we have counts for
10 papers · 1 filter
CycleFlow: Purify Information Factors by Cycle Loss
Haoran Sun, Chen Chen, Lantian Li +1
SpeechFlow is a powerful factorization model based on information bottleneck (IB), and its effectiveness has been reported by several studies. A potential problem of SpeechFlow, ho…
Neural Discriminant Analysis for Deep Speaker Embedding
Lantian Li, Dong Wang, Thomas Fang Zheng
Probabilistic Linear Discriminant Analysis (PLDA) is a popular tool in open-set classification/verification tasks. However, the Gaussian assumption underlying PLDA prevents it from…
ASR-Free Pronunciation Assessment
Sitong Cheng, Zhixin Liu, Lantian Li +3
Most of the pronunciation assessment methods are based on local features derived from automatic speech recognition (ASR), e.g., the Goodness of Pronunciation (GOP) score. In this p…
Domain-Invariant Speaker Vector Projection by Model-Agnostic Meta-Learning
Jiawen Kang, Ruiqi Liu, Lantian Li +3
Domain generalization remains a critical problem for speaker recognition, even with the state-of-the-art architectures based on deep neural nets. For example, a model trained on re…
Deep Normalization for Speaker Vectors
Yunqi Cai, Lantian Li, Dong Wang +1
Deep speaker embedding has demonstrated state-of-the-art performance in speaker recognition tasks. However, one potential issue with this approach is that the speaker vectors deriv…
CN-CELEB: a challenging Chinese speaker recognition dataset
Yue Fan, Jiawen Kang, Lantian Li +7
Recently, researchers set an ambitious goal of conducting speaker recognition in unconstrained conditions where the variations on ambient, channel and emotion could be arbitrary. H…