![](https://img-blog.csdnimg.cn/20201014180756927.png?x-oss-process=image/resize,m_fixed,h_64,w_64)
音频
CindyDawnmian
这个作者很懒,什么都没留下…
展开
-
提取mfcc,imfcc,cqcc,fft等各种语音特征
代码来自https://github.com/rosrad/asvspoof2017def trim_silence(audio, threshold=0.1, frame_length=2048): if audio.size < frame_length: frame_length = audio.size energy = librosa.feat...原创 2019-08-06 10:56:09 · 5552 阅读 · 6 评论 -
切割40s的音频 python
#coding:utf-8from pydub import AudioSegmentimport osinput_music = AudioSegment.from_wav("D:\dataset\\1.wav")output_music=AudioSegment.empty()start=float(0)dur=float(40*1000)output_music = inpu...原创 2019-09-19 12:04:14 · 209 阅读 · 0 评论