Recently, Xvectors and ECAPA-TDNN have been considered state-of-the-art models in designing speaker verification systems. This paper proposes a novel approach that combines Attentive statistic pooling-based Xvector and pre-trained ECAPA-TDNN for Vietnamese speaker verification. Experiments are conducted on various recent Vietnamese speech datasets.
Nội dung trích xuất từ tài liệu:
SV - VLSP 2021: Combine attentive statistical pooling based Xvector and pretrained ECAPA-TDNN for Vietnamese text-independent speaker verification
SV - VLSP 2021: Combine attentive statistical pooling based Xvector and pretrained ECAPA-TDNN for Vietnamese text-independent speaker verification
Số trang: 6
Loại file: pdf
Dung lượng: 462.15 KB
Lượt xem: 14
Lượt tải: 0
Thông tin tài liệu:
Tìm kiếm theo từ khóa liên quan:
Speaker verification Attentive statistical pooling Pre-trained ECAPA-TDNN Xvector model Multiple attention heads Distinguish speakersTài liệu có liên quan:
-
VLSP 2021 - SV challenge: Vietnamese speaker verification in noisy environments
8 trang 44 1 0 -
SV - VLSP2021: The smartcall - its's systems
5 trang 34 0 0
