Fulin Zhang — accepted AI-conference papers

2025

Codec-ASV: Exploring Neural Audio Codec For Speaker Representation Learning

Yuke Lin, Fulin Zhang, Yingying Gao, Shilei Zhang, Ming Li

ICASSP 2025accepted

Discrete speech representations have gained significant success in a variety of speech-related tasks. Among these, Neural Audio Codec (NAC), which serves as a compressed form of audio signals, have proven effective in speech AIGC applications. Moreover, we believe that the speaker information can be…

Cited by 0 Source ↗Scholar ↗

2025

Efficient Extreme Large-Scale Speaker Verification: Dynamic Active Sub Fully-Connected Layers for Faster Training and Memory Optimization

Fulin Zhang, Chenguang Hu, Yao Shen, Yingying Gao, Shilei Zhang, Junlan Feng

ICASSP 2025accepted

Using larger scale datasets in the training stage of speaker verification model usually leads to better performance. However, when the speaker number of the training dataset becomes extreme large (e.g., more than 1 million), the training speed and GPU memory demand will become bottlenecks which are…

Cited by 0 Source ↗Scholar ↗