BriefGPT.xyz
Jul, 2024
Papez: 基于听觉工作记忆的资源高效语音分离
Papez: Resource-Efficient Speech Separation with Auditory Working Memory
HTML
PDF
Hyunseok Oh, Juheon Yi, Youngki Lee
TL;DR
Papez是一种轻量级和计算高效的单通道语音分离模型，它通过使用小型听觉工作内存替代片段间的Transformer、自适应地修剪不需要进一步处理的输入令牌以及通过递归Transformer减少参数的方法，在资源和准确性方面取得了最好的平衡。
Abstract
transformer-based models
recently reached state-of-the-art single-channel
speech separation
accuracy; However, their extreme computational load makes it difficult to deploy them in resource-constrained mobile or
→