Research Article
PF-Net: Personalized Filter for Speaker Recognition from Raw Waveform
@INPROCEEDINGS{10.1007/978-3-031-23902-1_28, author={Wencheng Li and Zhenhua Tan and Zhenche Xia and Danke Wu and Jingyu Ning}, title={PF-Net: Personalized Filter for Speaker Recognition from Raw Waveform}, proceedings={Mobile Multimedia Communications. 15th EAI International Conference, MobiMedia 2022, Virtual Event, July 22-24, 2022, Proceedings}, proceedings_a={MOBIMEDIA}, year={2023}, month={2}, keywords={Speaker recognition Raw waveform Personalized filters Deep learning}, doi={10.1007/978-3-031-23902-1_28} }- Wencheng Li
Zhenhua Tan
Zhenche Xia
Danke Wu
Jingyu Ning
Year: 2023
PF-Net: Personalized Filter for Speaker Recognition from Raw Waveform
MOBIMEDIA
Springer
DOI: 10.1007/978-3-031-23902-1_28
Abstract
Speaker recognition using i-vector has been replaced by speaker recognition using deep learning. Speaker recognition based on Convolutional Neural Networks (CNNs) has been widely used in recent years, which learn low-level speech representations from raw waveforms. On this basis, a CNN architecture called SincNet proposes a kind of unique convolutional layer, which has achieved band-pass filters. Compared with standard CNNs, SincNet learns the low and high cut-off frequencies of each filter. This paper proposes an improved CNNs architecture called PF-Net, which encourages the first convolutional layer to implement more personalized filters than SincNet. PF-Net parameterizes the frequency domain shape and can realize band-pass filters by learning some deformation points in frequency domain. Compared with standard CNN, PF-Net can learn the characteristics of each filter. Compared with SincNet, PF-Net can learn more characteristic parameters, instead of only low and high cut-off frequencies. This provides a personalized filter bank for different tasks. As a result, our experiments show that the PF-Net converges faster than standard CNN and performs better than SincNet. Our code is available at github.com/TAN-OpenLab/PF-NET.
