def main():
false_missed_count = 7901
est_dir = "data/test/marks/"
ref_dir = "data/origin/cmu/cmu_us_bdl_arctic/marks/"
marks_extension = ".marks"
keys = set(file_names(ref_dir)) & set(file_names(est_dir))
total_count = 0
correct_count = 0
missed_count = 0
false_count = 0
errors = list()
for key in keys:
est_path = os.path.join(est_dir, key + marks_extension)
ref_path = os.path.join(ref_dir, key + marks_extension)
result_dict = evaluate(est_path, ref_path, radius=0.002)
total_count += result_dict["total_count"]
correct_count += result_dict["correct_count"]
missed_count += result_dict["missed_count"]
false_count += result_dict["false_count"]
errors.extend(result_dict["errors"])
"""revise false missed marks"""
total_count -= false_missed_count
missed_count -= false_missed_count
"""calculate rate"""
correct_rate = correct_count / total_count
missed_rate = missed_count / total_count
false_rate = false_count / total_count
std_value = std(errors)
acc_rate = accuracy(errors, mean=-0.0005)
print(
"correct_rate: {}, missed_rate: {}, false_rate: {}, std: {}, A25: {}.".
format(correct_rate, missed_rate, false_rate, std_value, acc_rate))
def make_mix2_data_set():
dst_path = "data/origin/cmu/mix2/"
wav_extension = ".wav"
marks_extension = ".marks"
"""jmk data set"""
jmk_path = "data/origin/cmu/cmu_us_jmk_arctic/"
jmk_train_keys, _ = random_keys(os.path.join(jmk_path, "marks"), num=400)
jmk_valid_keys = file_names(os.path.join(jmk_path, "validation_marks"))
copy_files(os.path.join(jmk_path, "wav"),
os.path.join(dst_path, "wav"),
jmk_train_keys,
wav_extension,
rename_prefix="jmk_")
copy_files(os.path.join(jmk_path, "wav"),
os.path.join(dst_path, "wav"),
jmk_valid_keys,
wav_extension,
rename_prefix="jmk_")
copy_files(os.path.join(jmk_path, "marks"),
os.path.join(dst_path, "marks"),
jmk_train_keys,
marks_extension,
rename_prefix="jmk_")
copy_files(os.path.join(jmk_path, "validation_marks"),
os.path.join(dst_path, "validation_marks"),
jmk_valid_keys,
marks_extension,
rename_prefix="jmk_")
"""bdl data set"""
bdl_path = "data/origin/cmu/cmu_us_bdl_arctic/"
bdl_train_keys, _ = random_keys(os.path.join(bdl_path, "marks"), num=800)
bdl_valid_keys, _ = random_keys(os.path.join(bdl_path, "validation_marks"),
num=30)
copy_files(os.path.join(bdl_path, "wav"),
os.path.join(dst_path, "wav"),
bdl_train_keys,
wav_extension,
rename_prefix="bdl_")
copy_files(os.path.join(bdl_path, "wav"),
os.path.join(dst_path, "wav"),
bdl_valid_keys,
wav_extension,
rename_prefix="bdl_")
copy_files(os.path.join(bdl_path, "marks"),
os.path.join(dst_path, "marks"),
bdl_train_keys,
marks_extension,
rename_prefix="bdl_")
copy_files(os.path.join(bdl_path, "validation_marks"),
os.path.join(dst_path, "validation_marks"),
bdl_valid_keys,
marks_extension,
rename_prefix="bdl_")
def main():
tf.logging.set_verbosity(tf.logging.INFO)
args = get_args()
tf.logging.info("Load wav from {}".format(args.wav_src_dir))
tf.logging.info("Generate marks to {}".format(args.marks_dst_dir))
net = GenerateModel()
graph = tf.Graph()
with graph.as_default():
with tf.variable_scope("data"):
inputs = tf.placeholder(tf.float32, shape=(None, crop_radius*2 + 1))
with tf.variable_scope("extract_model"):
tensor_dict = net.build(inputs, training=False)
predictions = tensor_dict["predictions"]
saver = tf.train.Saver()
config = tf.ConfigProto()
config.gpu_options.allow_growth = True
with tf.Session(graph=graph, config=config) as sess:
sess.run([tf.global_variables_initializer(), tf.local_variables_initializer()])
save_path = os.path.join(args.save_path, net.name)
if not load_model(saver, sess, save_path):
sess.run([tf.global_variables_initializer(), tf.local_variables_initializer()])
global_step_eval = 0
keys = file_names(args.wav_src_dir)
testing_steps = len(keys)
pbar = tqdm.tqdm(total=testing_steps)
pbar.update(global_step_eval)
for key in keys:
tf.logging.debug("File names: {}".format(key))
input_features = package_input_features(os.path.join(args.wav_src_dir, key + wav_extension))
predictions_eval = sess.run(predictions, feed_dict={inputs: input_features["wav"]})
marks = generate_marks(input_features["peak_indices"], predictions_eval, input_features["rate"])
tf.logging.debug("Number of marks: {}".format(len(marks)))
save_marks(os.path.join(args.marks_dst_dir, key + marks_extension), marks)
global_step_eval += 1
pbar.update(1)
tf.logging.info("Congratulations!")
# coding=utf-8
import os
import shutil
import random
from data_process.ops import file_names
validation_marks_path = "data/origin/cmu/cmu_us_slt_arctic/validation_marks/"
marks_path = "data/origin/cmu/cmu_us_slt_arctic/marks/"
wav_path = "data/origin/cmu/cmu_us_slt_arctic/wav/"
marks_keys = file_names(marks_path)
wav_keys = file_names(wav_path)
keys = list(set(marks_keys) - set(wav_keys))
print(keys)
for key in keys:
my_file = os.path.join(marks_path, key + ".marks")
os.remove(my_file)
marks_keys = file_names(marks_path)
intersection_keys = list(set(marks_keys) & set(wav_keys))
print(intersection_keys)
validation_keys = random.sample(intersection_keys, 100)
for key in validation_keys:
shutil.move(os.path.join(marks_path, key + ".marks"), os.path.join(validation_marks_path, key + ".marks"))
def random_keys(path, num=50):
all_keys = file_names(path)
training_keys = random.sample(all_keys, num)
testing_keys = list(set(all_keys) - set(training_keys))
return training_keys, testing_keys
def make_mix4_test_data_set():
dst_path = "data/origin/cmu/mix4/"
marks_extension = ".marks"
"""jmk data set"""
jmk_path = "data/origin/cmu/cmu_us_jmk_arctic/"
jmk_train_keys, jmk_test_keys = random_keys(os.path.join(
jmk_path, "marks"),
num=400)
jmk_valid_keys, _ = random_keys(os.path.join(jmk_path, "validation_marks"),
num=50)
copy_files(os.path.join(jmk_path, "marks"),
os.path.join(dst_path, "marks"),
jmk_train_keys,
marks_extension,
rename_prefix="jmk_")
copy_files(os.path.join(jmk_path, "validation_marks"),
os.path.join(dst_path, "validation_marks"),
jmk_valid_keys,
marks_extension,
rename_prefix="jmk_")
copy_files(os.path.join(jmk_path, "marks"),
os.path.join(dst_path, "jmk_testing_marks"),
jmk_test_keys,
marks_extension,
rename_prefix="jmk_")
"""bdl data set"""
bdl_path = "data/origin/cmu/cmu_us_bdl_arctic/"
bdl_train_keys, bdl_test_keys = random_keys(os.path.join(
bdl_path, "marks"),
num=800)
bdl_valid_keys = random_keys(os.path.join(bdl_path, "validation_marks"),
num=50)
copy_files(os.path.join(bdl_path, "marks"),
os.path.join(dst_path, "marks"),
bdl_train_keys,
marks_extension,
rename_prefix="bdl_")
copy_files(os.path.join(bdl_path, "validation_marks"),
os.path.join(dst_path, "validation_marks"),
bdl_valid_keys,
marks_extension,
rename_prefix="bdl_")
copy_files(os.path.join(bdl_path, "marks"),
os.path.join(dst_path, "bdl_testing_marks"),
bdl_test_keys,
marks_extension,
rename_prefix="bdl_")
"""slt data set"""
slt_path = "data/origin/cmu/cmu_us_slt_arctic/"
slt_train_keys, slt_test_keys = random_keys(os.path.join(
slt_path, "marks"),
num=400)
slt_valid_keys, _ = random_keys(os.path.join(slt_path, "validation_marks"),
num=50)
copy_files(os.path.join(slt_path, "marks"),
os.path.join(dst_path, "marks"),
slt_train_keys,
marks_extension,
rename_prefix="slt_")
copy_files(os.path.join(slt_path, "validation_marks"),
os.path.join(dst_path, "validation_marks"),
slt_valid_keys,
marks_extension,
rename_prefix="slt_")
copy_files(os.path.join(slt_path, "marks"),
os.path.join(dst_path, "slt_testing_marks"),
slt_test_keys,
marks_extension,
rename_prefix="slt_")
"""ked data set"""
ked_path = "data/origin/cmu/cmu_us_ked_timit/"
ked_train_keys, _ = random_keys(os.path.join(ked_path, "marks"), num=400)
ked_valid_keys = file_names(os.path.join(ked_path, "validation_marks"))
ked_test_keys = list(
set(all_keys(os.path.join(ked_path, "marks"))) - set(ked_train_keys))
copy_files(os.path.join(ked_path, "marks"),
os.path.join(dst_path, "marks"),
ked_train_keys,
marks_extension,
rename_prefix="ked_")
copy_files(os.path.join(ked_path, "validation_marks"),
os.path.join(dst_path, "validation_marks"),
ked_valid_keys,
marks_extension,
rename_prefix="ked_")
copy_files(os.path.join(ked_path, "marks"),
os.path.join(dst_path, "ked_testing_marks"),
ked_test_keys,
marks_extension,
rename_prefix="ked_")
def all_keys(path):
keys = file_names(path)
return keys