yongxuUSTC
commented
4 years ago Hi
This is according to the original noise aware training used in the robust ASR [1], which claims that "This performance can be further improved by incorporating information about the environment into DNN training using a new method called noise-aware training."
[1] An investigation of deep neural networks for noise robust speech recognition https://ieeexplore.ieee.org/abstract/document/6639100
On Mon, 23 Mar 2020 at 10:10, qq274943639 notifications@github.com wrote:
您好，我看了徐勇2015年的博士毕业论文，里面噪声告知训练的部分，在连续7帧即（7,257）的样本中通过取连续几帧的带噪语音的平均功率作为估计的平均噪声功率，这个依据是什么有些看不明白这部分的内容
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/yongxuUSTC/sednn/issues/54, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJGHURGLV5IDCFLHZPM7L3RI6JXVANCNFSM4LSA7NOQ .
您好，我看了徐勇2015年的博士毕业论文，里面噪声告知训练的部分，在连续7帧即（7,257）的样本中通过取连续几帧的带噪语音的平均功率作为估计的平均噪声功率，这个依据是什么有些看不明白这部分的内容