求翻译：在音频分类阶段，首先对音频进行预加重和分帧等预处理，接着提取出音频在时域和频域的3个特征：低能量帧率、最大带宽和基音周期可信度标准差，形成特征向量。然后采用支持向量机方法对找到音乐和语音这两类的最优分类面，建立模型，最后带入测试集检验分类效果。是什么意思？

Then extracted three characteristics of the audio in the time domain and frequency domain: low-energy frame rate, the maximum bandwidth and pitch credibility standards in audio classification stage, the first audio pre-emphasis and framing pretreatment, the formation of eigenvectors. And support vec

On the Audio classification stage, the first thing the audio for pre-emphasis and sub-frame, pre-processing, and then extract the audio from the time domain and frequency domain of the three characteristics: low-energy frame rate, maximum bandwidth and EUM cycle credibility standard deviation, a fea

In the audio frequency classification stage, first carries on pretreatments and so on pre-emphasis and minute frame to the audio frequency, then withdraws the audio frequency in the time domain and the frequency range 3 characteristics: The low energy frame rate, the maximum band width and the tone

Classification of audio stages, first to pre-add audio and framed, pre, then extracts 3 audio in time domain and frequency domain characteristics: low frame rate of energy and confidence of pitch standard deviation, maximum bandwidth, forming the characteristic vector. Then using support vector mach

正在翻译，请等待...