def feature_repository(cls, filename=None):
from dcase_util.processors import ProcessingChain
if filename is None:
filename = cls.audio_filename()
fs = 44100
chain = ProcessingChain([{
'processor_name':
'dcase_util.processors.MonoAudioReadingProcessor',
'init_parameters': {
'fs': fs
}
}, {
'processor_name':
'dcase_util.processors.RepositoryFeatureExtractorProcessor',
'init_parameters': {
'parameters': {
'mel': {},
'mfcc': {},
'mfcc_delta': {},
'mfcc_acceleration': {},
'zcr': {},
'rmse': {},
'centroid': {},
}
}
}])
return chain.process(filename=filename)
def process(self, data=None, store_processing_chain=False, **kwargs):
"""Extract features
Parameters
----------
data : AudioContainer
Audio data to extract features
store_processing_chain : bool
Store processing chain to data container returned
Default value False
Returns
-------
FeatureContainer
"""
from dcase_util.containers import FeatureContainer, AudioContainer
if isinstance(data, AudioContainer):
if store_processing_chain:
if hasattr(data, 'processing_chain') and data.processing_chain.chain_item_exists(
processor_name='dcase_util.processors.' + self.__class__.__name__):
# Current processor is already in the processing chain, get that
processing_chain_item = data.processing_chain.chain_item(
processor_name='dcase_util.processors.' + self.__class__.__name__
)
else:
# Create a new processing chain item
processing_chain_item = self.get_processing_chain_item()
processing_chain_item.update({
'process_parameters': kwargs
})
if hasattr(data, 'processing_chain'):
data.processing_chain.push_processor(**processing_chain_item)
processing_chain = data.processing_chain
else:
processing_chain = ProcessingChain().push_processor(**processing_chain_item)
else:
processing_chain = None
return FeatureContainer(
data=self.extract(y=data.get_focused()),
time_resolution=self.hop_length_seconds,
processing_chain=processing_chain
)
else:
message = '{name}: Wrong input data type, type required [{input_type}].'.format(
name=self.__class__.__name__,
input_type=self.input_type)
self.logger.exception(message)
raise ValueError(message)
def __init__(self, *args, **kwargs):
self.processing_chain = kwargs.get('processing_chain',
ProcessingChain())
if not hasattr(self, 'input_type'):
self.input_type = None
if not hasattr(self, 'output_type'):
self.output_type = None
def process(self, data=None, store_processing_chain=False, **kwargs):
"""Encode metadata
Parameters
----------
data : MetaDataContainer
Meta data to encode.
store_processing_chain : bool
Store processing chain to data container returned
Default value False
Returns
-------
BinaryMatrixContainer
"""
from dcase_util.containers import MetaDataContainer
if isinstance(data, MetaDataContainer):
# Do processing
self.encoder.encode(metadata_container=data)
if store_processing_chain:
# Get processing chain item
processing_chain_item = self.get_processing_chain_item()
processing_chain_item.update({'process_parameters': kwargs})
# Create processing chain to be stored in the container, and push chain item into it
if hasattr(data, 'processing_chain'):
data.processing_chain.push_processor(
**processing_chain_item)
processing_chain = data.processing_chain
else:
processing_chain = ProcessingChain().push_processor(
**processing_chain_item)
else:
processing_chain = None
from dcase_util.containers import BinaryMatrix2DContainer
container = BinaryMatrix2DContainer(
data=self.encoder.data,
label_list=self.encoder.label_list,
time_resolution=self.encoder.time_resolution,
processing_chain=processing_chain)
return container
else:
message = '{name}: Wrong input data type, type required [{input_type}].'.format(
name=self.__class__.__name__, input_type=self.input_type)
self.logger.exception(message)
raise ValueError(message)
def process(self,
data=None,
filename=None,
focus_start_samples=None,
focus_stop_samples=None,
focus_duration_samples=None,
focus_start_seconds=None,
focus_stop_seconds=None,
focus_duration_seconds=None,
focus_channel=None,
store_processing_chain=False,
**kwargs):
"""Audio reading
Parameters
----------
data :
filename : str
Filename
focus_start_samples : int
Sample index of focus segment start
focus_stop_samples : int
Sample index of focus segment stop
focus_duration_samples : int
Sample count of focus segment
focus_start_seconds : float
Time stamp (in seconds) of focus segment start
focus_stop_seconds : float
Time stamp (in seconds) of focus segment stop
focus_duration_seconds : float
Duration (in seconds) of focus segment
focus_channel : int or str
Audio channel id or name to focus. In case of stereo signal, valid channel labels to select
single channel are 'L', 'R', 'left', and 'right' or 0, 1, and to get mixed down version
of all channels 'mixdown'.
store_processing_chain : bool
Store processing chain to data container returned
Default value False
Returns
-------
AudioContainer
"""
if data is None and self.input_type == ProcessingChainItemType.NONE:
audio_container = AudioContainer(**self.init_parameters)
if filename:
audio_container.load(filename=filename,
mono=self.init_parameters.get('mono'))
# Set focus segment and channel
audio_container.set_focus(start=focus_start_samples,
stop=focus_stop_samples,
duration=focus_duration_samples,
start_seconds=focus_start_seconds,
stop_seconds=focus_stop_seconds,
duration_seconds=focus_duration_seconds,
channel=focus_channel)
if store_processing_chain:
processing_chain_item = self.get_processing_chain_item()
if 'process_parameters' not in processing_chain_item:
processing_chain_item['process_parameters'] = {}
processing_chain_item['process_parameters'][
'filename'] = filename
processing_chain_item['process_parameters'][
'focus_start_samples'] = focus_start_samples
processing_chain_item['process_parameters'][
'focus_stop_samples'] = focus_stop_samples
processing_chain_item['process_parameters'][
'focus_duration_samples'] = focus_duration_samples
processing_chain_item['process_parameters'][
'focus_start_seconds'] = focus_start_seconds
processing_chain_item['process_parameters'][
'focus_stop_seconds'] = focus_stop_seconds
processing_chain_item['process_parameters'][
'focus_duration_seconds'] = focus_duration_seconds
processing_chain_item['process_parameters'][
'focus_channel'] = focus_channel
# Push chain item into processing chain stored in the container
# Create processing chain to be stored in the container, and push chain item into it
if hasattr(audio_container, 'processing_chain'):
audio_container.processing_chain.push_processor(
**processing_chain_item)
else:
audio_container.processing_chain = ProcessingChain(
).push_processor(**processing_chain_item)
return audio_container
else:
message = '{name}: Wrong input data type, type required [{input_type}].'.format(
name=self.__class__.__name__, input_type=self.input_type)
self.logger.exception(message)
raise ValueError(message)
def process(self, data=None, store_processing_chain=False, **kwargs):
"""Extract features
Parameters
----------
data : AudioContainer
Audio data to extract features
store_processing_chain : bool
Store processing chain to data container returned
Default value False
Returns
-------
FeatureRepository
"""
from dcase_util.containers import FeatureRepository, AudioContainer
if isinstance(data, AudioContainer):
if store_processing_chain:
if hasattr(data, 'processing_chain'
) and data.processing_chain.chain_item_exists(
processor_name='dcase_util.processors.' +
self.__class__.__name__):
# Current processor is already in the processing chain, get that
processing_chain_item = data.processing_chain.chain_item(
processor_name='dcase_util.processors.' +
self.__class__.__name__)
else:
# Create a new processing chain item
processing_chain_item = self.get_processing_chain_item()
# Update current processing parameters into chain item
processing_chain_item.update({'process_parameters': kwargs})
# Create processing chain to be stored in the container, and push chain item into it
if hasattr(data, 'processing_chain'):
data.processing_chain.push_processor(
**processing_chain_item)
processing_chain = data.processing_chain
else:
processing_chain = ProcessingChain().push_processor(
**processing_chain_item)
else:
processing_chain = None
# Create repository container
repository = FeatureRepository(processing_chain=processing_chain)
# Make local copy of data
current_data = copy.deepcopy(data)
if data.streams == 1:
# We have single channel audio input
for label, parameters in iteritems(self.parameters):
if label in self.label_to_class:
# Get processor
processor = self.label_to_class[label](**parameters)
# Reset processing chain
current_data.processing_chain = ProcessingChain()
# Extract features
extracted = processor.process(data=current_data)
repository.set_container(container=extracted,
label=label)
else:
message = '{name}: Unknown label [{label}], no corresponding class found.'.format(
name=self.__class__.__name__, label=label)
self.logger.exception(message)
raise AssertionError(message)
elif data.streams > 1:
# We have multi-channel audio input
for stream_id in range(0, data.streams):
for label, parameters in iteritems(self.parameters):
if label in self.label_to_class:
# Get processor
processor = self.label_to_class[label](
**parameters)
# Reset processing chain
current_data.processing_chain = ProcessingChain()
# Set focus to the current stream
current_data.focus_channel = stream_id
# Extract features
extracted = processor.process(data=current_data)
# Add extracted features to the repository
repository.set_container(container=extracted,
label=label,
stream_id=stream_id)
else:
message = '{name}: Unknown label [{label}], no corresponding class found.'.format(
name=self.__class__.__name__, label=label)
self.logger.exception(message)
raise AssertionError(message)
return repository
def process(self,
data=None,
focus_field=None,
length_frames=None,
length_seconds=None,
store_processing_chain=False,
**kwargs):
"""Encode metadata
Parameters
----------
data : MetaDataContainer
Meta data to encode.
focus_field : str
Field from the meta data item to be used in encoding. If None, one given as parameter for
class constructor is used.
length_frames : int
Length of encoded segment in frames. If None, one given as parameter for class constructor is used.
length_seconds : float > 0.0
Length of encoded segment in seconds. If None, one given as parameter for class constructor is used.
store_processing_chain : bool
Store processing chain to data container returned
Default value False
Returns
-------
BinaryMatrixContainer
"""
from dcase_util.containers import MetaDataContainer
if focus_field is None:
focus_field = self.focus_field
if isinstance(data, MetaDataContainer):
if length_frames is None and length_seconds is not None:
length_frames = self._length_to_frames(length_seconds)
if length_frames is None:
length_frames = self.length_frames
if len(data) > 0:
label_list = data[0].get(focus_field)
if isinstance(label_list, str):
label_list = [label_list]
# Do processing
self.encode(label_list=label_list, length_frames=length_frames)
if store_processing_chain:
if hasattr(data, 'processing_chain'
) and data.processing_chain.chain_item_exists(
processor_name='dcase_util.processors.' +
self.__class__.__name__):
# Current processor is already in the processing chain, get that
processing_chain_item = data.processing_chain.chain_item(
processor_name='dcase_util.processors.' +
self.__class__.__name__)
else:
# Create a new processing chain item based on current processor class
processing_chain_item = self.get_processing_chain_item()
if 'process_parameters' not in processing_chain_item:
processing_chain_item['process_parameters'] = {}
processing_chain_item['process_parameters'][
'focus_field'] = focus_field
processing_chain_item['process_parameters'][
'length_frames'] = length_frames
# Create processing chain to be stored in the container, and push chain item into it
if hasattr(data, 'processing_chain'):
data.processing_chain.push_processor(
**processing_chain_item)
processing_chain = data.processing_chain
else:
processing_chain = ProcessingChain().push_processor(
**processing_chain_item)
else:
processing_chain = None
from dcase_util.containers import BinaryMatrix2DContainer
container = BinaryMatrix2DContainer(
data=self.data,
label_list=self.label_list,
time_resolution=self.time_resolution,
processing_chain=processing_chain)
return container
else:
message = '{name}: Wrong input data type, type required [{input_type}].'.format(
name=self.__class__.__name__, input_type=self.input_type)
self.logger.exception(message)
raise ValueError(message)
def process(self,
data=None,
label=None,
focus_field=None,
length_frames=None,
length_seconds=None,
store_processing_chain=False,
**kwargs):
"""Encode metadata
Parameters
----------
data : MetaDataContainer
Meta data to encode. Give data in either through meta data container or directly with label parameter.
label : str
Class label to be hot
focus_field : str
Field from the meta data item to be used in encoding. If None, one given as parameter for class
constructor is used.
length_frames : int
Length of encoded segment in frames. If None, one given as parameter for class constructor is used.
length_seconds : float > 0.0
Length of encoded segment in seconds. If None, one given as parameter for class constructor is used.
store_processing_chain : bool
Store processing chain to data container returned
Default value False
Returns
-------
BinaryMatrixContainer
"""
if data is None and label is None:
message = '{name}: Give data or label parameter.'.format(
name=self.__class__.__name__)
self.logger.exception(message)
raise ValueError(message)
from dcase_util.containers import MetaDataContainer
if data is not None and not isinstance(data, MetaDataContainer):
message = '{name}: Wrong input data type, type required [{input_type}].'.format(
name=self.__class__.__name__, input_type=self.input_type)
self.logger.exception(message)
raise ValueError(message)
if focus_field is None:
focus_field = self.focus_field
if data is not None and len(data) > 0 and label is None:
label = data[0].get(focus_field)
# Do processing
self.encoder.encode(label=label,
length_frames=length_frames,
length_seconds=length_seconds)
if store_processing_chain:
# Get processing chain item
processing_chain_item = self.get_processing_chain_item()
if 'process_parameters' not in processing_chain_item:
processing_chain_item['process_parameters'] = {}
processing_chain_item['process_parameters'][
'focus_field'] = focus_field
processing_chain_item['process_parameters'][
'length_frames'] = length_frames
# Create processing chain to be stored in the container, and push chain item into it
if hasattr(data, 'processing_chain'):
data.processing_chain.push_processor(**processing_chain_item)
processing_chain = data.processing_chain
else:
processing_chain = ProcessingChain().push_processor(
**processing_chain_item)
else:
processing_chain = None
from dcase_util.containers import BinaryMatrix2DContainer
container = BinaryMatrix2DContainer(
data=self.encoder.data,
label_list=self.encoder.label_list,
time_resolution=self.encoder.time_resolution,
processing_chain=processing_chain)
return container