Chrome Extension
WeChat Mini Program
Use on ChatGLM

基于NMF和FCRF的单通道语音分离

Journal of Tsinghua University(2017)

Cited 1|Views3
No score
Abstract
近年来,非负矩阵分解(non-negative matrix factorization, NMF)被广泛应用于单通道语音分离问题.然而,标准的NMF算法假设语音的相邻帧之间是相互独立的,不能表征语音信号的时间连续性信息.为此,该文提出了一种基于NMF和因子条件随机场(factorial conditional random field,FCRF)的语音分离算法,首先将NMF和k均值聚类结合对纯净语音的频谱结构以及时间连续性进行建模,然后利用得到的模型训练FCRF模型,进而对混合语音信号进行分离.结果表明:该算法相比没有考虑语音时间连续特性的基于NMF 的算法如激活集牛顿算法(active-set Newton algorithm,ASNA),在客观指标上有明显提高.
More
Translated text
Key words
single-channel speech separation,factorial conditional random field (FCRF),non-negative matrix factorization (NMF),k-means clustering
AI Read Science
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined