Listening tests show that the similarity between target speech and modified one reached 78.8%.
听力测试表明,转换后的语音和目标说话人的语音相似度达到78.8%。
This paper proposes a new speech enhancement scheme based on modified Minimum Mean Square Error (MMSE) estimation of speech Short Time Spectrum Amplitude (STSA).
本文提出了一种改进型语音短时谱最小均方误差(MMSE)估计的增强方法。
A modified MMSE algorithm for speech enhancement based on speech periodic property is presented.
基于语音周期性的特点,提出了一种预加重的MMSE语音增强的改进算法。
应用推荐