分别使用GMM 和 HMM 建立对 0～9 十个数字的孤立词语音分类识别模型，给出python代码

这里给出使用GMM和HMM进行语音分类识别的Python代码示例，包括数据预处理、模型训练和测试三个部分。代码中使用了Scikit-learn和Python_speech_features两个库来进行特征提取、模型训练和测试。

1. 数据预处理

import os
import numpy as np
import scipy.io.wavfile as wav
from python_speech_features import mfcc
from sklearn.model_selection import train_test_split

# 定义数据路径
data_path = 'data/'

# 定义数字和标签
digits = ['0', '1', '2', '3', '4', '5', '6', '7', '8', '9']
labels = [0, 1, 2, 3, 4, 5, 6, 7, 8, 9]

# 定义mfcc参数
numcep = 13
winlen = 0.025
winstep = 0.01

# 定义空列表来存储特征和标签
X = []
y = []

# 遍历数据路径，加载数据并提取特征
for label, digit in enumerate(digits):
    digit_path = os.path.join(data_path, digit)
    for filename in os.listdir(digit_path):
        filepath = os.path.join(digit_path, filename)
        rate, signal = wav.read(filepath)
        mfcc_feat = mfcc(signal, rate, numcep=numcep, winlen=winlen, winstep=winstep)
        X.append(mfcc_feat)
        y.append(label)

# 将特征和标签转化为numpy数组并进行训练集和测试集的划分
X = np.array(X)
y = np.array(y)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

2. GMM模型训练和测试

from sklearn.mixture import GaussianMixture

# 定义GMM模型
n_components = 16
cov_type = 'diag'
gmm = GaussianMixture(n_components=n_components, covariance_type=cov_type)

# 训练GMM模型
X_train_flat = np.vstack(X_train)
gmm.fit(X_train_flat)

# 测试GMM模型
X_test_flat = np.vstack(X_test)
y_pred_gmm = gmm.predict(X_test_flat)

# 计算GMM模型的准确率
accuracy_gmm = np.mean(y_pred_gmm == y_test)
print('GMM accuracy:', accuracy_gmm)

3. HMM模型训练和测试

from hmmlearn import hmm

# 定义HMM模型
n_components = 16
cov_type = 'diag'
hmm_model = hmm.GaussianHMM(n_components=n_components, covariance_type=cov_type)

# 训练HMM模型
X_train_lens = [len(x) for x in X_train]
X_train_flat = np.vstack(X_train)
hmm_model.fit(X_train_flat, X_train_lens)

# 测试HMM模型
X_test_lens = [len(x) for x in X_test]
X_test_flat = np.vstack(X_test)
log_likelihoods = hmm_model.score_samples(X_test_flat)
y_pred_hmm = np.argmax(log_likelihoods, axis=1)

# 计算HMM模型的准确率
accuracy_hmm = np.mean(y_pred_hmm == y_test)
print('HMM accuracy:', accuracy_hmm)

需要注意的是，上述代码仅为简单示例，实际应用中需要进行更多的参数调整和模型优化。同时，也需要根据实际数据情况进行适当的调整。