数据集和代码均已上传到Github中，欢迎大家下载使用。
Github地址：https://github.com/JasonZhang156/Sound-Recognition-Tutorial
如果这个教程对您有所帮助，请不吝贡献您的小星星Q^Q.
构建模型
【声音识别|4. 构建模型】本节使用keras搭建一个简单的CNN模型。该CNN模型包括3个卷积层、3个池化层、2个全连接层，中间层激活函数使用ReLU，最后一层使用softmax，每个卷积层后使用 Batch Normalization加速训练。优化器使用SGD，损失函数使用交叉熵（Cross Entropy）。模型详细配置如下：
文章图片
Keras实现代码如下：
# -*- coding: utf-8 -*-
"""
@author: Jason Zhang
@github: https://github.com/JasonZhang156/Sound-Recognition-Tutorial
"""from keras.layers import Input
from keras.layers import Conv2D, MaxPool2D, Dense, Dropout, BatchNormalization, Activation, GlobalAvgPool2D
from keras.models import Model
from keras import optimizers
from keras.utils import plot_modeldef CNN(input_shape=(60,65,1), nclass=10):
"""
build a simple cnn model using keras with TensorFlow backend.
:param input_shape: input shape of network, default as (60,65,1)
:param nclass: numbers of class(output shape of network), default as 10
:return: cnn model
"""
input_ = Input(shape=input_shape)# Conv1
x = Conv2D(64, kernel_size=(3, 3), strides=(1, 1), padding='same')(input_)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = MaxPool2D(pool_size=(2, 2), strides=(2, 2))(x)# Conv2
x = Conv2D(128, kernel_size=(3, 3), strides=(1, 1), padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = MaxPool2D(pool_size=(2, 2), strides=(2, 2))(x)# Conv3
x = Conv2D(256, kernel_size=(3, 3), strides=(1, 1), padding='same')(x)
x = BatchNormalization()(x)
x = Activation('relu')(x)
x = MaxPool2D(pool_size=(2, 2), strides=(2, 2))(x)# GAP
x = GlobalAvgPool2D()(x)
# Dense_1
x = Dense(256, activation='relu')(x)
x = Dropout(0.5)(x)
# Dense_2
output_ = Dense(nclass, activation='softmax')(x)model = Model(inputs=input_, outputs=output_)
# 输出模型的参数信息
model.summary()
# 配置模型训练过程
sgd = optimizers.sgd(lr=0.01, momentum=0.9, nesterov=True)# 优化器为SGD
model.compile(loss='categorical_crossentropy', optimizer=sgd, metrics=['accuracy'])# 交叉熵为cross entropyreturn modelif __name__ == '__main__':
model = CNN()
plot_model(model, './image/cnn.png')# 保存模型图上述代码包含在Github中的models.py文件中。
