8 citations · 14 across the 11 of their papers we have counts for
11 papers
StatQAT: Statistical Quantizer Optimization for Deep Networks
Mehmet Aktukmak, Daniel Huang, Ke Ding
Quantization is essential for reducing the computational cost and memory usage of deep neural networks, enabling efficient inference on low-precision hardware. Despite the growing…
Learning Reduced-Order Models for Cardiovascular Simulations with Graph Neural Networks
Luca Pegolotti, Martin R. Pfaller, Natalia L. Rubio +4
Reduced-order models based on physics are a popular choice in cardiovascular modeling due to their efficiency, but they may experience reduced accuracy when working with anatomies…
Improve Noise Tolerance of Robust Loss via Noise-Awareness
Kehui Ding, Jun Shu, Deyu Meng +1
Robust loss minimization is an important strategy for handling robust learning issue on noisy labels. Current approaches for designing robust losses involve the introduction of noi…
Label-free Knowledge Distillation with Contrastive Loss for Light-weight Speaker Recognition
Zhiyuan Peng, Xuanji He, Ke Ding +2
Very deep models for speaker recognition (SR) have demonstrated remarkable performance improvement in recent research. However, it is impractical to deploy these models for on-devi…
Covariance Regularization for Probabilistic Linear Discriminant Analysis
Zhiyuan Peng, Mingjie Shao, Xuanji He +4
Probabilistic linear discriminant analysis (PLDA) is commonly used in speaker verification systems to score the similarity of speaker embeddings. Recent studies improved the perfor…
Peak-First CTC: Reducing the Peak Latency of CTC Models by Applying Peak-First Regularization
Zhengkun Tian, Hongyu Xiang, Min Li +3
The CTC model has been widely applied to many application scenarios because of its simple structure, excellent performance, and fast inference speed. There are many peaks in the pr…
Unifying Cosine and PLDA Back-ends for Speaker Verification
Zhiyuan Peng, Xuanji He, Ke Ding +2
State-of-art speaker verification (SV) systems use a back-end model to score the similarity of speaker embeddings extracted from a neural network model. The commonly used back-end…
An Empirical Study of Language Model Integration for Transducer based Speech Recognition
Huahuan Zheng, Keyu An, Zhijian Ou +3
Utilizing text-only data with an external language model (ELM) in end-to-end RNN-Transducer (RNN-T) for speech recognition is challenging. Recently, a class of methods such as dens…
CUSIDE: Chunking, Simulating Future Context and Decoding for Streaming ASR
Keyu An, Huahuan Zheng, Zhijian Ou +3
History and future contextual information are known to be important for accurate acoustic modeling. However, acquiring future context brings latency for streaming ASR. In this pape…
Dynamic-TinyBERT: Boost TinyBERT's Inference Efficiency by Dynamic Sequence Length
Shira Guskin, Moshe Wasserblat, Ke Ding +1
Limited computational budgets often prevent transformers from being used in production and from having their high accuracy utilized. TinyBERT addresses the computational efficiency…
Learning Speaker Embedding with Momentum Contrast
Ke Ding, Xuanji He, Guanglu Wan
Speaker verification can be formulated as a representation learning task, where speaker-discriminative embeddings are extracted from utterances of variable lengths. Momentum Contra…