基于keras的中文语音识别
- 该项目实现了GRU-CTC中文语音识别，所有代码都在
gru_ctc_am.py中，包括：- 音频文件特征提取
- 文本数据处理
- 数据格式处理
- 构建模型
- 模型训练及解码
- 之外还包括将aishell数据处理为thchs30数据格式，合并数据进行训练。代码及数据放在
gen_aishell_data中。
默认数据集为thchs30，参考gen_aishell_data中的数据及代码，也可以使用aishell的数据进行训练。
音频文件特征提取
# -----------------------------------------------------------------------------------------------------
'''
&usage: [audio]对音频文件进行处理，包括生成总的文件列表、特征提取等
'''
# -----------------------------------------------------------------------------------------------------
# 生成音频列表
def genwavlist(wavpath):
wavfiles = {}
fileids = []
for (dirpath, dirnames, filenames) in os.walk(wavpath):
for filename in filenames:
if filename.endswith('.wav'):
filepath = os.sep.join([dirpath, filename])
fileid = filename.strip('.wav')
wavfiles[fileid] = filepath
fileids.append(fileid)
return wavfiles,fileids
# 对音频文件提取mfcc特征
def compute_mfcc(file):
fs, audio = wav.read(file)
mfcc_feat = mfcc(audio, samplerate=fs, numcep=26)
mfcc_feat = mfcc_feat[::3]
mfcc_feat = np.transpose(mfcc_feat)
mfcc_feat = pad_sequences(mfcc_feat, maxlen=500, dtype='float', padding='post', truncating='post').T
return mfcc_feat
文本数据处理
# -----------------------------------------------------------------------------------------------------
'''
&usage: [text]对文本标注文件进行处理，包括生成拼音到数字的映射，以及将拼音标注转化为数字的标注转化
'''
# -----------------------------------------------------------------------------------------------------
# 利用训练数据生成词典
def gendict(textfile_path):
dicts = []
textfile = open(textfile_path,'r+')
for content in textfile.readlines():
content = content.strip('\n')
content = content.split(' ',1)[1]
content = content.split(' ')
dicts += (word for word in content)
counter = Counter(dicts)
words = sorted(counter)
wordsize = len(words)
word2num = dict(zip(words, range(wordsize)))
num2word = dict(zip(range(wordsize), words))
return word2num, num2word #1176个音素
# 文本转化为数字
def text2num(textfile_path):
lexcion,num2word = gendict(textfile_path)
word2num = lambda word:lexcion.get(word, 0)
textfile = open(textfile_path, 'r+')
content_dict = {}
for content in textfile.readlines():
content = content.strip('\n')
cont_id = content.split(' ',1)[0]
content = content.split(' ',1)[1]
content = content.split(' ')
content = list(map(word2num,content))
add_num = list(np.zeros(50-len(content)))
content = content + add_num
content_dict[cont_id] = content
return content_dict,lexcion
数据格式处理
# -----------------------------------------------------------------------------------------------------
'''
&usage: [data]数据生成器构造，用于训练的数据生成，包括输入特征及标注的生成，以及将数据转化为特定格式
'''
# -----------------------------------------------------------------------------------------------------
# 将数据格式整理为能够被网络所接受的格式，被data_generator调用
def get_batch(x, y, train=False, max_pred_len=50, input_length=500):
X = np.expand_dims(x, axis=3)
X = x # for model2
# labels = np.ones((y.shape[0], max_pred_len)) * -1 # 3 # , dtype=np.uint8
labels = y
input_length = np.ones([x.shape[0], 1]) * ( input_length - 2 )
# label_length = np.ones([y.shape[0], 1])
label_length = np.sum(labels > 0, axis=1)
label_length = np.expand_dims(label_length,1)
inputs = {'the_input': X,
'the_labels': labels,
'input_length': input_length,
'label_length': label_length,
}
outputs = {'ctc': np.zeros([x.shape[0]])} # dummy data for dummy loss function
return (inputs, outputs)
# 数据生成器，默认音频为thchs30\train,默认标注为thchs30\train.syllable,被模型训练方法fit_generator调用
def data_generate(wavpath = 'E:\\Data\\data_thchs30\\train', textfile = 'E:\\Data\\thchs30\\train.syllable.txt', bath_size=4):
wavdict,fileids = genwavlist(wavpath)
#print(wavdict)
content_dict,lexcion = text2num(textfile)
genloop = len(fileids)//bath_size
print("all loop :", genloop)
while True:
feats = []
labels = []
# 随机选择某个音频文件作为训练数据
i = random.randint(0,genloop-1)
for x in range(bath_size):
num = i * bath_size + x
fileid = fileids[num]
# 提取音频文件的特征
mfcc_feat = compute_mfcc(wavdict[fileid])
feats.append(mfcc_feat)
# 提取标注对应的label值
labels.append(content_dict[fileid])
# 将数据格式修改为get_batch可以处理的格式
feats = np.array(feats)
labels = np.array(labels)
# 调用get_batch将数据处理为训练所需的格式
inputs, outputs = get_batch(feats, labels)
yield inputs, outputs
构建模型
# -----------------------------------------------------------------------------------------------------
'''
&usage: [net model]构件网络结构，用于最终的训练和识别
'''
# -----------------------------------------------------------------------------------------------------
# 被creatModel调用，用作ctc损失的计算
def ctc_lambda(args):
labels, y_pred, input_length, label_length = args
y_pred = y_pred[:, :, :]
return K.ctc_batch_cost(labels, y_pred, input_length, label_length)
# 构建网络结构，用于模型的训练和识别
def creatModel():
input_data = Input(name='the_input', shape=(500, 26))
layer_h1 = Dense(512, activation="relu", use_bias=True, kernel_initializer='he_normal')(input_data)
#layer_h1 = Dropout(0.3)(layer_h1)
layer_h2 = Dense(512, activation="relu", use_bias=True, kernel_initializer='he_normal')(layer_h1)
layer_h3_1 = GRU(512, return_sequences=True, kernel_initializer='he_normal', dropout=0.3)(layer_h2)
layer_h3_2 = GRU(512, return_sequences=True, go_backwards=True, kernel_initializer='he_normal', dropout=0.3)(layer_h2)
layer_h3 = add([layer_h3_1, layer_h3_2])
layer_h4 = Dense(512, activation="relu", use_bias=True, kernel_initializer='he_normal')(layer_h3)
#layer_h4 = Dropout(0.3)(layer_h4)
layer_h5 = Dense(1177, activation="relu", use_bias=True, kernel_initializer='he_normal')(layer_h4)
output = Activation('softmax', name='Activation0')(layer_h5)
model_data = Model(inputs=input_data, outputs=output)
#ctc
labels = Input(name='the_labels', shape=[50], dtype='float32')
input_length = Input(name='input_length', shape=[1], dtype='int64')
label_length = Input(name='label_length', shape=[1], dtype='int64')
loss_out = Lambda(ctc_lambda, output_shape=(1,), name='ctc')([labels, output, input_length, label_length])
model = Model(inputs=[input_data, labels, input_length, label_length], outputs=loss_out)
model.summary()
ada_d = Adadelta(lr=0.01, rho=0.95, epsilon=1e-06)
#model=multi_gpu_model(model,gpus=2)
model.compile(loss={'ctc': lambda y_true, output: output}, optimizer=ada_d)
#test_func = K.function([input_data], [output])
print("model compiled successful!")
return model, model_data
模型训练及解码
# -----------------------------------------------------------------------------------------------------
'''
&usage: 模型的解码，用于将数字信息映射为拼音
'''
# -----------------------------------------------------------------------------------------------------
# 对model预测出的softmax的矩阵，使用ctc的准则解码，然后通过字典num2word转为文字
def decode_ctc(num_result, num2word):
result = num_result[:, :, :]
in_len = np.zeros((1), dtype = np.int32)
in_len[0] = 50;
r = K.ctc_decode(result, in_len, greedy = True, beam_width=1, top_paths=1)
r1 = K.get_value(r[0][0])
r1 = r1[0]
text = []
for i in r1:
text.append(num2word[i])
return r1, text
# -----------------------------------------------------------------------------------------------------
'''
&usage: 模型的训练
'''
# -----------------------------------------------------------------------------------------------------
# 训练模型
def train():
# 准备训练所需数据
yielddatas = data_generate()
# 导入模型结构，训练模型，保存模型参数
model, model_data = creatModel()
model.fit_generator(yielddatas, steps_per_epoch=2000, epochs=1)
model.save_weights('model.mdl')
model_data.save_weights('model_data.mdl')
# -----------------------------------------------------------------------------------------------------
'''
&usage: 模型的测试，看识别结果是否正确
'''
# -----------------------------------------------------------------------------------------------------
# 测试模型
def test():
# 准备测试数据，以及生成字典
word2num, num2word = gendict('E:\\Data\\thchs30\\train.syllable.txt')
yielddatas = data_generate(bath_size=1)
# 载入训练好的模型，并进行识别
model, model_data = creatModel()
model_data.load_weights('model_data.mdl')
result = model_data.predict_generator(yielddatas, steps=1)
# 将数字结果转化为文本结果
result, text = decode_ctc(result, num2word)
print('数字结果： ', result)
print('文本结果：', text)
aishell数据转化
将aishell中的汉字标注转化为拼音标注，利用该数据与thchs30数据训练同样的网络结构。
该模型作为一个练手小项目。
没有使用语言模型，直接简单建模。