登录
洞察 - Acoustic Signal Processing
Efficient Sound Field Reconstruction using Conditional Invertible Neural Networks
A conditional invertible neural network (CINN) is proposed to efficiently reconstruct sound fields in reverberant environments, balancing accuracy and computational efficiency while incorporating uncertainty estimates.
GESI: Gammachirp Envelope Similarity Index for Predicting Speech Intelligibility of Simulated Hearing Loss Sounds
提案された客観的理解度尺度（OIM）であるGESIは、聴力損失音の実験を通じて正常聴力リスナーの音声理解度（SI）を予測することが可能であり、これにより補聴器向けのSEアルゴリズムの改善が可能となる。
Unsupervised Acoustic Scene Mapping Using RTF and LOCA
Utilizing RTF and LOCA for robust acoustic scene mapping in reverberant environments.
