基礎(chǔ)語(yǔ)音識(shí)別-食物語(yǔ)音識(shí)別baseline(CNN)

!wget http://tianchi-competition.oss-cn-hangzhou.aliyuncs.com/531887/train_sample.zip
!unzip -qq train_sample.zip
!\rm train_sample.zip
!wget http://tianchi-competition.oss-cn-hangzhou.aliyuncs.com/531887/test_a.zip
!unzip -qq test_a.zip
!\rm test_a.zip

# 基本庫(kù)
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split  #劃分?jǐn)?shù)據(jù)集
from sklearn.metrics import classification_report   #用于顯示主要分類指標(biāo)的文本報(bào)告
from sklearn.model_selection import GridSearchCV #自動(dòng)調(diào)參
from sklearn.preprocessing import MinMaxScaler #歸一化

# 搭建分類模型所需要的庫(kù)
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Conv2D, Flatten, Dense, MaxPool2D, Dropout
from tensorflow.keras.utils import to_categorical 
from sklearn.ensemble import RandomForestClassifier
from sklearn.svm import SVC #支持向量分類
!pip install librosa --user #加載音頻處理庫(kù)
# 其他庫(kù)
import os
import librosa #音頻處理庫(kù)
import librosa.display
import glob

feature = []
label = []
# 建立類別標(biāo)簽，不同類別對(duì)應(yīng)不同的數(shù)字。
label_dict = {'aloe': 0, 'burger': 1, 'cabbage': 2,'candied_fruits':3, 'carrots': 4, 'chips':5,
                  'chocolate': 6, 'drinks': 7, 'fries': 8, 'grapes': 9, 'gummies': 10, 'ice-cream':11,
                  'jelly': 12, 'noodles': 13, 'pickles': 14, 'pizza': 15, 'ribs': 16, 'salmon':17,
                  'soup': 18, 'wings': 19}
label_dict_inv = {v:k for k,v in label_dict.items()}

from tqdm import tqdm
def extract_features(parent_dir, sub_dirs, max_file=10, file_ext="*.wav"):
    c = 0
    label, feature = [], []
    for sub_dir in sub_dirs:
        for fn in tqdm(glob.glob(os.path.join(parent_dir, sub_dir, file_ext))[:max_file]): # 遍歷數(shù)據(jù)集的所有文件
           # segment_log_specgrams, segment_labels = [], []
            #sound_clip,sr = librosa.load(fn)
            #print(fn)
            label_name = fn.split('/')[-2]
            label.extend([label_dict[label_name]])
            X, sample_rate = librosa.load(fn,res_type='kaiser_fast')
            mels = np.mean(librosa.feature.melspectrogram(y=X,sr=sample_rate).T,axis=0) # 計(jì)算梅爾頻譜(mel spectrogram),并把它作為特征
            feature.extend([mels])
    return [feature, label]

# 自己更改目錄
parent_dir = './train_sample/'
save_dir = "./"
folds = sub_dirs = np.array(['aloe','burger','cabbage','candied_fruits',
                             'carrots','chips','chocolate','drinks','fries',
                            'grapes','gummies','ice-cream','jelly','noodles','pickles',
                            'pizza','ribs','salmon','soup','wings'])
# 獲取特征feature以及類別的label
temp = extract_features(parent_dir,sub_dirs,max_file=100)
temp = np.array(temp)
data = temp.transpose()

# 獲取特征
X = np.vstack(data[:, 0])
# 獲取標(biāo)簽
Y = np.array(data[:, 1])
print('X的特征尺寸是：',X.shape)
print('Y的特征尺寸是：',Y.shape)

# 在Keras庫(kù)中：to_categorical就是將類別向量轉(zhuǎn)換為二進(jìn)制（只有0和1）的矩陣類型表示
Y = to_categorical(Y)
print(X.shape)
print(Y.shape)

X_train, X_test, Y_train, Y_test = train_test_split(X, Y, random_state = 1, stratify=Y)
print('訓(xùn)練集的大小',len(X_train))
print('測(cè)試集的大小',len(X_test))

X_train = X_train.reshape(-1, 16, 8, 1)
X_test = X_test.reshape(-1, 16, 8, 1)

model = Sequential()
# 輸入的大小
input_dim = (16, 8, 1)
model.add(Conv2D(64, (3, 3), padding = "same", activation = "tanh", input_shape = input_dim))# 卷積層
model.add(MaxPool2D(pool_size=(2, 2)))# 最大池化
model.add(Conv2D(128, (3, 3), padding = "same", activation = "tanh")) #卷積層
model.add(MaxPool2D(pool_size=(2, 2))) # 最大池化層
model.add(Dropout(0.1))
model.add(Flatten()) # 展開(kāi)
model.add(Dense(1024, activation = "tanh"))
model.add(Dense(20, activation = "softmax")) # 輸出層：20個(gè)units輸出20個(gè)類的概率
# 編譯模型，設(shè)置損失函數(shù)，優(yōu)化方法以及評(píng)價(jià)標(biāo)準(zhǔn)
model.compile(optimizer = 'adam', loss = 'categorical_crossentropy', metrics = ['accuracy'])
model.summary()

# 訓(xùn)練模型
model.fit(X_train, Y_train, epochs = 100, batch_size = 15, validation_data = (X_test, Y_test))

def extract_features(test_dir, file_ext="*.wav"):
    feature = []
    for fn in tqdm(glob.glob(os.path.join(test_dir, file_ext))[:]): # 遍歷數(shù)據(jù)集的所有文件
        X, sample_rate = librosa.load(fn,res_type='kaiser_fast')
        mels = np.mean(librosa.feature.melspectrogram(y=X,sr=sample_rate).T,axis=0) # 計(jì)算梅爾頻譜(mel spectrogram),并把它作為特征
        feature.extend([mels])
    return feature
X_test = extract_features('./test_a/')
X_test = np.vstack(X_test)
predictions = model.predict(X_test.reshape(-1, 16, 8, 1))
preds = np.argmax(predictions, axis = 1)
preds = [label_dict_inv[x] for x in preds]
path = glob.glob('./test_a/*.wav')
result = pd.DataFrame({'name':path, 'label': preds})
result['name'] = result['name'].apply(lambda x: x.split('/')[-1])
result.to_csv('submit.csv',index=None)
!ls ./test_a/*.wav | wc -l
!wc -l submit.csv