Multi-Resolution Multi-Head Attention in Deep Speaker Embedding.

Zhiming Wang Kaisheng Yao Xiaolong Li Shuo Fang

Published in: ICASSP (2020)

Keyphrases

multiresolution
wavelet transform
vector space
neural network
subband
speaker recognition
speech recognition
coarse to fine
hierarchical representation
real time
data sets
genetic algorithm
feature extraction
hidden markov models
vector quantization
automatic speech recognition