dcase_util.containers.AudioContainer

class dcase_util.containers.AudioContainer(data=None, fs=44100, focus_start_samples=None, focus_stop_samples=None, focus_channel=None, channel_labels=None, **kwargs)

Audio container class.

Constructor

Parameters

datanumpy.ndarray or list of numpy.ndarray

Data to initialize the container Default value None

fsint

Target sampling frequency, if loaded audio does have different sampling frequency, audio will be re-sampled. Default value “44100”

focus_start_samplesint

Focus segment start Default value None

focus_stop_samplesint

Focus segment stop Default value None

focus_channelint

Focus segment channel Default value None

channel_labelslist

Channel names Default value None

filenamestr, optional

File path

__init__(data=None, fs=44100, focus_start_samples=None, focus_stop_samples=None, focus_channel=None, channel_labels=None, **kwargs)

Constructor

Parameters

datanumpy.ndarray or list of numpy.ndarray

Data to initialize the container Default value None

fsint

Target sampling frequency, if loaded audio does have different sampling frequency, audio will be re-sampled. Default value “44100”

focus_start_samplesint

Focus segment start Default value None

focus_stop_samplesint

Focus segment stop Default value None

focus_channelint

Focus segment channel Default value None

channel_labelslist

Channel names Default value None

filenamestr, optional

File path

Methods

__init__([data, fs, focus_start_samples, ...])	Constructor
delimiter([exclude_delimiters])	Use csv.sniffer to guess delimiter for CSV file
detect_file_format([filename])	Detect file format from extension
exists()	Checks that file exists
frames([frame_length, hop_length, ...])	Slice audio into overlapping frames.
freeze()	Freeze focus segment, copy segment to be container's data.
get_file_information()	Get file information, filename
get_focused()	Get focus segment from audio data.
is_package([filename])	Determine if the file is compressed package.
load([filename, fs, mono, res_type, start, ...])	Load file
load_from_youtube(query_id[, start, stop, ...])	Load audio data from youtube
log([level])	Log container content
mixdown()	Mix all audio channels into single channel.
normalize([headroom])	Normalize audio data.
overlay(audio[, start_seconds, multiplier])	Simple sample overlay method
pad([type, length, length_seconds])	Generate signal
plot([plot_type])	Visualize audio data
plot_spec([spec_type, hop_length, cmap, ...])	Visualize audio data as spectrogram.
plot_wave([x_axis, max_points, max_sr, ...])	Visualize audio data as waveform.
resample(target_fs[, scale, res_type])	Resample audio data.
reset_focus()	Reset focus segment.
save([filename, bit_depth, bit_rate])	Save audio
segments([segment_length, ...])	Slice audio into segments.
set_focus([start, stop, duration, ...])	Set focus segment
show([mode, indent, visualize])	Print container content
to_html([indent])	Get container information in a HTML formatted string
to_string([ui, indent])	Get container information in a string
validate_format()	Validate file format

Attributes

bytes	File size in bytes
channels	Number of audio channels.
data	Audio data
duration_ms	Duration of audio data in milliseconds.
duration_samples	Duration of audio data in samples.
duration_sec	Duration of audio data in seconds.
empty	Check if audio data is empty.
focus_channel	Focus channel
focus_start_samples	Focus segment start in samples.
focus_start_seconds	Focus segment start in seconds.
focus_stop_samples	Focus segment stop in samples.
focus_stop_seconds	Focus segment stop in seconds.
length	Length of audio data in samples.
loaded	Audio load status.
logger	Logger instance
md5	Checksum for file.
shape	Audio data shape.
streams	Rename channels for compatibility.
valid_formats	Valid file formats