Bi-Sep: A Multi-Resolution Cross-Domain Monaural Speech Separation Framework

Kuan Hsun Ho, Jeih Weih Hung, Berlin Chen

研究成果: 書貢獻/報告類型會議論文篇章

1 引文 斯高帕斯(Scopus)

摘要

In recent years, deep neural network (DNN)-based time-domain methods for monaural speech separation have substantially improved under an anechoic condition. However, the performance of these methods degrades when facing harsher conditions, such as noise or reverberation. Although adopting Short-Time Fourier Transform (STFT) for feature extraction of these neural methods helps stabilize the performance in non-anechoic situations, it inherently loses the fine-grained vision, which is one of the particularities of time-domain methods. Therefore, this study explores incorporating time and STFT-domain features to retain their beneficial characteristics. Furthermore, we leverage a Bi-Projection Fusion (BPF) mechanism to merge the information between two domains. To evaluate the effectiveness of our proposed method, we conduct experiments in an anechoic setting on the WSJ0-2mix dataset and noisy/reverberant settings on WHAM!/WHAMR! dataset. The experiment shows that with a cost of ignorable degradation on anechoic dataset, the proposed method manages to promote the performance of existing neural models when facing more complicated environments.

原文英語
主出版物標題Proceedings - 2022 International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022
發行者Institute of Electrical and Electronics Engineers Inc.
頁面72-77
頁數6
ISBN(電子)9798350399509
DOIs
出版狀態已發佈 - 2022
事件27th International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022 - Tainan, 臺灣
持續時間: 2022 12月 12022 12月 3

出版系列

名字Proceedings - 2022 International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022

會議

會議27th International Conference on Technologies and Applications of Artificial Intelligence, TAAI 2022
國家/地區臺灣
城市Tainan
期間2022/12/012022/12/03

ASJC Scopus subject areas

  • 人工智慧
  • 電腦科學應用
  • 硬體和架構
  • 控制和優化

指紋

深入研究「Bi-Sep: A Multi-Resolution Cross-Domain Monaural Speech Separation Framework」主題。共同形成了獨特的指紋。

引用此