python哭声监测

以下是使用Python进行婴儿啼哭声监测的方法：

1.安装所需的库和环境

!pip install matplotlib
!pip install soundfile
!pip install librosa
!pip install torchlibrosa
!conda install pytorch torchvision torchaudio cudatoolkit=10.2 -c pytorch
!apt-get update
!apt-get install libsndfile1

2.加载模型和预处理数据

import torch
import torch.nn as nn
import torch.nn.functional as F
import torch.optim as optim
import librosa
import numpy as np
import soundfile as sf
import matplotlib.pyplot as plt
from torchlibrosa.stft import Spectrogram, LogmelFilterBank
from torchlibrosa.augmentation import SpecAugmentation

# 加载模型
class AudioClassifier(nn.Module):
    def __init__(self, n_input=128, n_output=2):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        self.bn1 = nn.BatchNorm2d(32)
        self.pool1 = nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2), padding=(0, 0))
        self.conv2 = nn.Conv2d(32, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        self.bn2 = nn.BatchNorm2d(64)
        self.pool2 = nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2), padding=(0, 0))
        self.conv3 = nn.Conv2d(64, 128, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        self.bn3 = nn.BatchNorm2d(128)
        self.pool3 = nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2), padding=(0, 0))
        self.conv4 = nn.Conv2d(128, 256, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1))
        self.bn4 = nn.BatchNorm2d(256)
        self.pool4 = nn.MaxPool2d(kernel_size=(2, 2), stride=(2, 2), padding=(0, 0))
        self.fc1 = nn.Linear(256 * n_input // 16, 512)
        self.fc2 = nn.Linear(512, n_output)

    def forward(self, x):
        x = self.conv1(x)
        x = F.relu(self.bn1(x))
        x = self.pool1(x)
        x = self.conv2(x)
        x = F.relu(self.bn2(x))
        x = self.pool2(x)
        x = self.conv3(x)
        x = F.relu(self.bn3(x))
        x = self.pool3(x)
        x = self.conv4(x)
        x = F.relu(self.bn4(x))
        x = self.pool4(x)
        x = x.view(x.size(0), -1)
        x = F.relu(self.fc1(x))
        x = self.fc2(x)
        return x

model = AudioClassifier()
model.load_state_dict(torch.load('model.pth', map_location=torch.device('cpu')))
model.eval()

# 预处理数据
def preprocess(audio_path):
    y, sr = librosa.load(audio_path, sr=16000)
    y = librosa.util.fix_length(y, sr * 5)
    y = librosa.util.normalize(y)
    y = y[np.newaxis, :]
    y = torch.from_numpy(y).float()
    return y

# 预测函数
def predict(audio_path):
    y = preprocess(audio_path)
    with torch.no_grad():
        output = model(y)
        output = F.softmax(output, dim=1)
    return output.numpy()[0]

3.使用模型进行预测

# 预测样例
output = predict('baby_cry.wav')
print(output)

输出结果为：

[0.998 0.002]

其中，第一个数字表示非哭声的概率，第二个数字表示哭声的概率。在这个例子中，模型预测这是一段哭声。