咨询与建议

限定检索结果

文献类型

  • 1 篇 期刊文献

馆藏范围

  • 1 篇 电子文献
  • 0 种 纸本馆藏

日期分布

学科分类号

  • 1 篇 理学
    • 1 篇 物理学
  • 1 篇 工学
    • 1 篇 电气工程

主题

  • 1 篇 speech dereverbe...
  • 1 篇 noise reduction
  • 1 篇 floating-point i...
  • 1 篇 task analysis
  • 1 篇 noise measuremen...
  • 1 篇 speech enhanceme...
  • 1 篇 inference algori...
  • 1 篇 computational mo...
  • 1 篇 inference accele...
  • 1 篇 deep neural netw...
  • 1 篇 signal processin...

机构

  • 1 篇 johns hopkins un...
  • 1 篇 natl taiwan univ...
  • 1 篇 city univ hong k...
  • 1 篇 acad sinica res ...

作者

  • 1 篇 fu szu-wei
  • 1 篇 tsao yu
  • 1 篇 kuo tei-wei
  • 1 篇 yu cheng
  • 1 篇 hsu yi-te
  • 1 篇 lin yu-chen

语言

  • 1 篇 英文
检索条件"主题词=floating-point integer arithmetic circuit"
1 条 记 录,以下是1-10 订阅
排序:
SEOFP-NET: Compression and Acceleration of Deep Neural Networks for Speech Enhancement Using Sign-Exponent-Only floating-points
收藏 引用
IEEE-ACM TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING 2022年 30卷 1016-1031页
作者: Lin, Yu-Chen Yu, Cheng Hsu, Yi-Te Fu, Szu-Wei Tsao, Yu Kuo, Tei-Wei Acad Sinica Res Ctr Informat Technol Innovat Taipei 11529 Taiwan Johns Hopkins Univ Dept Comp Sci Baltimore MD 21287 USA Natl Taiwan Univ Dept Comp Sci & Informat Engn Taipei 10617 Taiwan City Univ Hong Kong Coll Engn Kowloon Tong Hong Kong Peoples R China
Numerous compression and acceleration strategies have achieved outstanding results on classification tasks in various fields. Nevertheless, the same strategies may yield unsatisfactory performance on regression tasks ... 详细信息
来源: 评论