Sound Event Localization and Detection Based on Time-Frequency Separable Convolutional Compression Network

Shih Tsung Yang, Fong Ci Jhou, Jia Ching Wang, Pao Chi Chang

研究成果: 書貢獻/報告類型會議論文篇章同行評審

摘要

This work proposes a Time-Frequency Separable Convolutional Compression Network (TFSCCN) as a system architecture for sound event localization and detection. It utilizes 1-D convolution kernels of different dimensions to extract features of time and frequency components separately, and also reduces the amount of model parameters by controlling the increase or decrease of the number of channels in the neural network. In addition, the model combines multi-head self-attention (MHSA) to obtain global and local information in time series features, and uses dual-branch tracking technology to effectively locate and detect the same or different overlapping sound events.

原文???core.languages.en_GB???
主出版物標題2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021
發行者Institute of Electrical and Electronics Engineers Inc.
頁面432-433
頁數2
ISBN(電子)9781665436762
DOIs
出版狀態已出版 - 2021
事件10th IEEE Global Conference on Consumer Electronics, GCCE 2021 - Kyoto, Japan
持續時間: 12 10月 202115 10月 2021

出版系列

名字2021 IEEE 10th Global Conference on Consumer Electronics, GCCE 2021

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???10th IEEE Global Conference on Consumer Electronics, GCCE 2021
國家/地區Japan
城市Kyoto
期間12/10/2115/10/21

指紋

深入研究「Sound Event Localization and Detection Based on Time-Frequency Separable Convolutional Compression Network」主題。共同形成了獨特的指紋。

引用此