Introduction

語音識別(speech recognition)是將語音內容轉換為對應文字的技術。

連結時間序列分類(Connectionist Temporal Classification, CTC)，是時間序列(seq2seq)的一種訓練方法，使用遞歸神經網路標記序列資料。

Tasks

引用物件。

import os
import cntk as C
import numpy as np

import cntk.tests.test_utils
cntk.tests.test_utils.set_device_from_pytest_env()

1.資料讀取(Data reading)：

下載資料至本地端資料夾。

data_dir = os.path.join("..", "Tests", "EndToEndTests", "Speech", "Data")
print("Current directory {0}".format(os.getcwd()))

if os.path.exists(data_dir):
    if os.path.realpath(data_dir) != os.path.realpath(os.getcwd()):
        os.chdir(data_dir)
        print("Changed to data directory {0}".format(data_dir))
else:
    print("Data directory not available locally. Downloading data.")
    try:
        from urllib.request import urlretrieve
    except ImportError:
        from urllib import urlretrieve
    for dir in ['GlobalStats', 'Features']:
        if not os.path.exists(dir):
            os.mkdir(dir)
    for file in ['glob_0000.scp', 'glob_0000.write.scp', 'glob_0000.mlf', 'state_ctc.list', 'GlobalStats/mean.363', 'GlobalStats/var.363', 'Features/000000000.chunk']:
        if os.path.exists(file):
            print('Already downloaded %s' % file)
        else:
            print('Downloading %s' % file)
            urlretrieve('https://github.com/Microsoft/CNTK/raw/release/2.3.1/Tests/EndToEndTests/Speech/Data/%s' % file, file)

HTK/MLF格式的聲學模型(Acoustic Model, AM)訓練資料集：

• SCP：發音編號(utterance ids)及相應的特徵值(feature)。
• MLF：主標籤檔案(master label file, MLF)，錄音(transcription)的特徵值(features)。
• States list file：標籤列表檔案。

讀取資料，使用 CNTK 的 HTK 資料讀取器，HTKFeatureDeserializer、HTKMLFDeserializer。

feature_dimension = 33
feature = C.sequence.input((feature_dimension))

label_dimension = 133
label = C.sequence.input((label_dimension))

train_feature_filepath = "glob_0000.scp"
train_label_filepath = "glob_0000.mlf"
mapping_filepath = "state_ctc.list"
try:
    train_feature_stream = C.io.HTKFeatureDeserializer(
    C.io.StreamDefs(speech_feature = C.io.StreamDef(shape = feature_dimension, scp = train_feature_filepath)))
    train_label_stream = C.io.HTKMLFDeserializer(
    mapping_filepath, C.io.StreamDefs(speech_label = C.io.StreamDef(shape = label_dimension, mlf = train_label_filepath)), True)
    train_data_reader = C.io.MinibatchSource([train_feature_stream, train_label_stream], frame_mode = False)
    train_input_map = {feature: train_data_reader.streams.speech_feature, label: train_data_reader.streams.speech_label}
except RuntimeError:
    print ("ERROR: not able to read features or labels")

2.資料處理(Data preprocessing)：

將輸入的特徵值減去平均值，乘以標準差的倒數(reciprocal)，將其標準化。

feature_mean = np.fromfile(os.path.join("GlobalStats", "mean.363"), dtype=float, count=feature_dimension)
feature_inverse_stddev = np.fromfile(os.path.join("GlobalStats", "var.363"), dtype=float, count=feature_dimension)

feature_normalized = (feature - feature_mean) * feature_inverse_stddev

with C.default_options(activation=C.sigmoid):
    z = C.layers.Sequential([
        C.layers.For(range(3), lambda: C.layers.Recurrence(C.layers.LSTM(1024))),
        C.layers.Dense(label_dimension)
    ])(feature_normalized)

3.建立模型(Model creation)：

CTC準則函數通過結合labels_to_graph函數和

定義超參數。
labels_to_graph，轉換輸入的標籤序列，成為圖形結構資料。
forward_backward，歸納時間序列經常使用的雙向學習速率。

mbsize = 1024
mbs_per_epoch = 10
max_epochs = 5

criteria = C.forward_backward(C.labels_to_graph(label), z, blankTokenId=132, delayConstraint=3)
err = C.edit_distance_error(z, label, squashInputs=True, tokensToIgnore=[132])

lr = C.learning_parameter_schedule_per_sample([(3, .01), (1,.001)])
mm = C.momentum_schedule([(1000, 0.9), (0, 0.99)], mbsize)
learner = C.momentum_sgd(z.parameters, lr, mm)
trainer = C.Trainer(z, (criteria, err), learner)

4.訓練模型(Learning the model)：

C.logging.log_number_of_parameters(z)
progress_printer = C.logging.progress_print.ProgressPrinter(tag='Training', num_epochs = max_epochs)

for epoch in range(max_epochs):
    for mb in range(mbs_per_epoch):
        minibatch = train_data_reader.next_minibatch(mbsize, input_map = train_input_map)
        trainer.train_minibatch(minibatch)
        progress_printer.update_with_trainer(trainer, with_metric = True)

    print('Trained on a total of ' + str(trainer.total_number_of_samples_seen) + ' frames')
    progress_printer.epoch_summary(with_metric = True)

# 儲存模型
# z.save('CTC_' + str(max_epochs) + 'epochs_' + str(mbsize) + 'mbsize_' + str(mbs_per_epoch) + 'mbs.model')

5.評估模型(Evaluation)：

test_feature_filepath = "glob_0000.write.scp"
test_feature_stream = C.io.HTKFeatureDeserializer(
    C.io.StreamDefs(speech_feature = C.io.StreamDef(shape = feature_dimension, scp = test_feature_filepath)))
test_data_reader = C.io.MinibatchSource([test_feature_stream, train_label_stream], frame_mode = False)
test_input_map = {feature: test_data_reader.streams.speech_feature, label: test_data_reader.streams.speech_label}

num_test_minibatches = 2
test_result = 0.0
for i in range(num_test_minibatches):
    test_minibatch = test_data_reader.next_minibatch(mbsize, input_map = test_input_map)
    eval_error = trainer.test_minibatch(test_minibatch)
    test_result = test_result + eval_error

# 評估誤差的平均值
round(test_result / num_test_minibatches,2)