PATH_ACOUSTIC_FEAT, PATH_EXP)
# ===========================================================================
# Config
# ===========================================================================
stdio(os.path.join(PATH_EXP, 'features_extraction.log'))
args = args_parse(descriptions=[
('recipe', 'the name of function defined in feature_recipes.py', None),
('--debug', 'enable debug or not', None, False)
])
DEBUG = args.debug
# ===========================================================================
# Create the recipes
# ===========================================================================
extractor = get_module_from_path(identifier=str(args.recipe),
prefix='feature_recipes',
path=get_script_path())
assert len(extractor) > 0, \
"Cannot find any recipe with name: '%s' from path: '%s'" % (args.recipe, get_script_path())
recipe = extractor[0](DEBUG)
# ====== debugging ====== #
if DEBUG:
with np.warnings.catch_warnings():
np.warnings.filterwarnings('ignore')
for path, name in SAMPLED_WAV_FILE:
feat = recipe.transform(path)
assert feat['bnf'].shape[0] == feat['mspec'].shape[0]
V.plot_multiple_features(feat, title=feat['name'])
V.plot_save(os.path.join(PATH_EXP, 'features_%s.pdf' % args.recipe))
exit()
# ===========================================================================
# Prepare the processor
from utils import (WAV_FILES, SAMPLED_WAV_FILE, PATH_ACOUSTIC_FEAT, PATH_EXP)
# ===========================================================================
# Config
# ===========================================================================
stdio(os.path.join(PATH_EXP, 'features_extraction.log'))
args = args_parse(
descriptions=[('recipe',
'the name of function defined in feature_recipes.py',
None), ('--debug', 'enable debug or not', None, False)])
DEBUG = args.debug
# ===========================================================================
# Create the recipes
# ===========================================================================
extractor = get_module_from_path(identifier=str(args.recipe),
prefix='feature_recipes',
path=get_script_path())
assert len(extractor) > 0, \
"Cannot find any recipe with name: '%s' from path: '%s'" % (args.recipe, get_script_path())
recipe = extractor[0](DEBUG)
# ====== debugging ====== #
if DEBUG:
with np.warnings.catch_warnings():
np.warnings.filterwarnings('ignore')
for path, name in SAMPLED_WAV_FILE:
feat = recipe.transform(path)
assert feat['bnf'].shape[0] == feat['mspec'].shape[0]
V.plot_multiple_features(feat, title=feat['name'])
V.plot_save(os.path.join(PATH_EXP, 'features_%s.pdf' % args.recipe))
exit()
# ===========================================================================
# Prepare the processor
catch_warnings_ignore)
from odin import fuel as F, preprocessing as pp
from odin.stats import sampling_iter
from helpers import (PATH_ACOUSTIC_FEATURES, EXP_DIR, ALL_FILES, IS_DEBUGGING,
FEATURE_RECIPE, FEATURE_NAME, ALL_DATASET, Config, NCPU,
validate_features_dataset)
# ALL_FILES
# Header:
# 0 1 2 3 4 5 6
# path, channel, name, spkid, dataset, start_time, end_time
# ===========================================================================
# Extractor
# ===========================================================================
recipe = get_module_from_path(identifier=FEATURE_RECIPE,
path=get_script_path(),
prefix='feature_recipes')
if len(recipe) == 0:
raise ValueError("Cannot find feature recipe with name: '%s'" %
FEATURE_RECIPE)
recipe = recipe[0]()
# ===========================================================================
# Debug mode
# ===========================================================================
if IS_DEBUGGING:
with np.warnings.catch_warnings():
rand = np.random.RandomState(seed=Config.SUPER_SEED)
np.warnings.filterwarnings('ignore')
# ====== stratify sampling from each dataset ====== #
clusters = defaultdict(list)
clusters_count = defaultdict(int)
# ====== check the running script to determine the current running states ====== #
_script_name = get_script_name()
if _script_name in ('speech_augmentation', 'speech_features_extraction'):
CURRENT_STATE = SystemStates.EXTRACT_FEATURES
_check_feature_extraction_requirement()
_check_recipe_name_for_extraction()
elif _script_name in ('train_xvec', 'train_ivec', 'train_tvec', 'train_evec',
'analyze', 'analyze_data'):
CURRENT_STATE = SystemStates.TRAINING
elif _script_name in ('make_score'):
CURRENT_STATE = SystemStates.SCORING
_check_feature_extraction_requirement()
else:
raise RuntimeError("Unknown states for current running script: %s/%s" %
(get_script_path(), get_script_name()))
# some fancy log of current state
print(ctext('====================================', 'red'))
print(ctext("System state:", 'cyan'), ctext(CURRENT_STATE, 'yellow'))
print(ctext('====================================', 'red'))
# ===========================================================================
# FILE LIST PATH
# ===========================================================================
# ====== basic directories ====== #
EXP_DIR = get_exppath('sre', override=False)
# this folder store extracted vectors for training backend and extracting scores
VECTORS_DIR = os.path.join(EXP_DIR, 'vectors')
if not os.path.exists(VECTORS_DIR):
os.mkdir(VECTORS_DIR)
# this folder store the results
RESULT_DIR = os.path.join(EXP_DIR, 'results')
DATA_DIR)
else:
DATA_DIR = select_path(os.path.join(DEFAULT_BASE_DIR, 'bio_data'),
create_new=True)
DOWNLOAD_DIR = select_path(os.path.join(DATA_DIR, 'downloads'),
create_new=True)
# PATH for saving experiments results
if 'SISUA_EXP' in os.environ:
EXP_DIR = os.environ['SISUA_EXP']
if not os.path.exists(EXP_DIR):
os.mkdir(EXP_DIR)
elif os.path.isfile(EXP_DIR):
raise RuntimeError("Experiment path at '%s' must be a folder" %
EXP_DIR)
else:
EXP_DIR = select_path(os.path.join(DEFAULT_BASE_DIR, 'bio_exp'),
create_new=True)
# ====== path for yaml configurations ====== #
if 'SISUA_CFG' in os.environ:
CONFIG_PATH = os.path.abspath(os.environ['SISUA_CFG'])
else:
CONFIG_PATH = os.path.abspath(
os.path.join(get_script_path(__name__, return_dir=True), '..', '..',
'configs', 'base.yaml'))
if not os.path.isfile(CONFIG_PATH):
raise RuntimeError("Cannot find configuration .yaml files at: %s" %
CONFIG_PATH)
from odin import fuel as F, preprocessing as pp
from odin.stats import sampling_iter
from helpers import (PATH_ACOUSTIC_FEATURES, EXP_DIR,
ALL_FILES, IS_DEBUGGING, FEATURE_RECIPE, FEATURE_NAME,
ALL_DATASET, Config, NCPU,
validate_features_dataset)
# ALL_FILES
# Header:
# 0 1 2 3 4 5 6
# path, channel, name, spkid, dataset, start_time, end_time
# ===========================================================================
# Extractor
# ===========================================================================
recipe = get_module_from_path(identifier=FEATURE_RECIPE,
path=get_script_path(),
prefix='feature_recipes')
if len(recipe) == 0:
raise ValueError("Cannot find feature recipe with name: '%s'" % FEATURE_RECIPE)
recipe = recipe[0]()
# ===========================================================================
# Debug mode
# ===========================================================================
if IS_DEBUGGING:
with np.warnings.catch_warnings():
rand = np.random.RandomState(seed=Config.SUPER_SEED)
np.warnings.filterwarnings('ignore')
# ====== stratify sampling from each dataset ====== #
clusters = defaultdict(list)
clusters_count = defaultdict(int)
samples = []
raise ValueError("'_' can appear in recipe name which is: '%s'" % FEATURE_RECIPE)
# ====== check the running script to determine the current running states ====== #
_script_name = get_script_name()
if _script_name in ('speech_augmentation', 'speech_features_extraction'):
CURRENT_STATE = SystemStates.EXTRACT_FEATURES
_check_feature_extraction_requirement()
_check_recipe_name_for_extraction()
elif _script_name in ('train_xvec', 'train_ivec', 'train_tvec',
'train_evec', 'analyze', 'analyze_data'):
CURRENT_STATE = SystemStates.TRAINING
elif _script_name in ('make_score'):
CURRENT_STATE = SystemStates.SCORING
_check_feature_extraction_requirement()
else:
raise RuntimeError("Unknown states for current running script: %s/%s" %
(get_script_path(), get_script_name()))
# some fancy log of current state
print(ctext('====================================', 'red'))
print(ctext("System state:", 'cyan'), ctext(CURRENT_STATE, 'yellow'))
print(ctext('====================================', 'red'))
# ===========================================================================
# FILE LIST PATH
# ===========================================================================
# ====== basic directories ====== #
EXP_DIR = get_exppath('sre', override=False)
# this folder store extracted vectors for training backend and extracting scores
VECTORS_DIR = os.path.join(EXP_DIR, 'vectors')
if not os.path.exists(VECTORS_DIR):
os.mkdir(VECTORS_DIR)
# this folder store the results
RESULT_DIR = os.path.join(EXP_DIR, 'results')