Mercurial > hg > auditok
changeset 404:08a7af37f2e9
Update code with new pre-commit hooks
| author | Amine Sehili <amine.sehili@gmail.com> |
|---|---|
| date | Sun, 26 May 2024 23:29:33 +0200 |
| parents | 996948ada980 |
| children | f56b4d8adfb8 |
| files | auditok/__init__.py auditok/cmdline.py auditok/cmdline_util.py auditok/core.py auditok/io.py auditok/plotting.py auditok/signal.py auditok/signal_numpy.py auditok/workers.py doc/conf.py setup.py |
| diffstat | 11 files changed, 143 insertions(+), 140 deletions(-) [+] |
line wrap: on
line diff
--- a/auditok/__init__.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/__init__.py Sun May 26 23:29:33 2024 +0200 @@ -10,8 +10,8 @@ """ from .core import * +from .exceptions import * from .io import * from .util import * -from .exceptions import * __version__ = "0.2.0"
--- a/auditok/cmdline.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/cmdline.py Sun May 26 23:29:33 2024 +0200 @@ -14,19 +14,19 @@ @deffield updated: 01 Mar 2021 """ +import os import sys -import os +import threading +import time from argparse import ArgumentParser -import time -import threading -from auditok import __version__, AudioRegion +from auditok import AudioRegion, __version__ + +from . import workers +from .cmdline_util import initialize_workers, make_kwargs, make_logger +from .exceptions import AudioEncodingWarning, EndOfProcessing +from .io import player_for from .util import AudioDataSource -from .exceptions import EndOfProcessing, AudioEncodingWarning -from .io import player_for -from .cmdline_util import make_logger, make_kwargs, initialize_workers -from . import workers - __all__ = [] __date__ = "2015-11-23"
--- a/auditok/cmdline_util.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/cmdline_util.py Sun May 26 23:29:33 2024 +0200 @@ -1,9 +1,10 @@ +import logging import sys -import logging from collections import namedtuple + from . import workers -from .util import AudioDataSource from .io import player_for +from .util import AudioReader _AUDITOK_LOGGER = "AUDITOK_LOGGER" KeywordArguments = namedtuple( @@ -81,7 +82,7 @@ def initialize_workers(logger=None, **kwargs): observers = [] - reader = AudioDataSource(source=kwargs["input"], **kwargs) + reader = AudioReader(source=kwargs["input"], **kwargs) if kwargs["save_stream"] is not None: reader = workers.StreamSaverWorker( reader,
--- a/auditok/core.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/core.py Sun May 26 23:29:33 2024 +0200 @@ -7,11 +7,14 @@ AudioRegion StreamTokenizer """ + +import math import os -import math -from .util import AudioReader, DataValidator, AudioEnergyValidator -from .io import check_audio_data, to_file, player_for, get_audio_source -from .exceptions import TooSamllBlockDuration + +from .exceptions import TooSmallBlockDuration +from .io import check_audio_data, get_audio_source, player_for, to_file +from .plotting import plot +from .util import AudioEnergyValidator, AudioReader, DataValidator try: from . import signal_numpy as signal @@ -90,7 +93,7 @@ max_silence=0.3, drop_trailing_silence=False, strict_min_dur=False, - **kwargs + **kwargs, ): """ Split audio data and return a generator of AudioRegions @@ -104,17 +107,17 @@ Every object that is not an `AudioReader` will be transformed into an `AudioReader` before processing. If it is an `str` that refers to a raw audio file, `bytes` or None, audio parameters should be provided using - kwargs (i.e., `samplig_rate`, `sample_width` and `channels` or their + kwargs (i.e., `sampling_rate`, `sample_width` and `channels` or their alias). If `input` is str then audio format will be guessed from file extension. `audio_format` (alias `fmt`) kwarg can also be given to specify audio format explicitly. If none of these options is available, rely on backend (currently only pydub is supported) to load data. min_dur : float, default: 0.2 - minimun duration in seconds of a detected audio event. By using large + minimum duration in seconds of a detected audio event. By using large values for `min_dur`, very short audio events (e.g., very short 1-word - utterances like 'yes' or 'no') can be mis detected. Using very short - values might result in a high number of short, unuseful audio events. + utterances like 'yes' or 'no') can be mis detected. Using a very small + value may result in a high number of too short audio events. max_dur : float, default: 5 maximum duration in seconds of a detected audio event. If an audio event lasts more than `max_dur` it will be truncated. If the continuation of a @@ -177,7 +180,7 @@ max_read, mr : float, default: None, read until end of stream maximum data to read from source in seconds. validator, val : callable, DataValidator - custom data validator. If `None` (default), an `AudioEnergyValidor` is + custom data validator. If `None` (default), an `AudioEnergyValidtor` is used with the given energy threshold. Can be a callable or an instance of `DataValidator` that implements `is_valid`. In either case, it'll be called with with a window of audio data as the first parameter. @@ -197,11 +200,11 @@ a generator of detected :class:`AudioRegion` s. """ if min_dur <= 0: - raise ValueError("'min_dur' ({}) must be > 0".format(min_dur)) + raise ValueError(f"'min_dur' ({min_dur}) must be > 0") if max_dur <= 0: - raise ValueError("'max_dur' ({}) must be > 0".format(max_dur)) + raise ValueError(f"'max_dur' ({max_dur}) must be > 0") if max_silence < 0: - raise ValueError("'max_silence' ({}) must be >= 0".format(max_silence)) + raise ValueError(f"'max_silence' ({max_silence}) must be >= 0") if isinstance(input, AudioReader): source = input @@ -212,7 +215,7 @@ ) if analysis_window <= 0: raise ValueError( - "'analysis_window' ({}) must be > 0".format(analysis_window) + f"'analysis_window' ({analysis_window}) must be > 0" ) params = kwargs.copy() @@ -225,11 +228,12 @@ input = bytes(input) try: source = AudioReader(input, block_dur=analysis_window, **params) - except TooSamllBlockDuration as exc: - err_msg = "Too small 'analysis_windows' ({0}) for sampling rate " - err_msg += "({1}). Analysis windows should at least be 1/{1} to " - err_msg += "cover one single data sample" - raise ValueError(err_msg.format(exc.block_dur, exc.sampling_rate)) + except TooSmallBlockDuration as exc: + err_msg = f"Too small 'analysis_window' ({exc.block_dur}) for " + err_msg += f"sampling rate ({exc.sampling_rate}). Analysis window " + err_msg += f"should at least be 1/{exc.sampling_rate} to cover " + err_msg += "one data sample" + raise ValueError(err_msg) from exc validator = kwargs.get("validator", kwargs.get("val")) if validator is None: @@ -358,8 +362,8 @@ frame_duration: float duration of analysis window in seconds start_frame : int - index of the fisrt analysis window - samling_rate : int + index of the first analysis window + sampling_rate : int sampling rate of audio data sample_width : int number of bytes of one audio sample @@ -369,7 +373,7 @@ Returns ------- audio_region : AudioRegion - AudioRegion whose start time is calculeted as: + AudioRegion whose start time is calculated as: `1000 * start_frame * frame_duration` """ start = start_frame * frame_duration @@ -648,13 +652,15 @@ @property def seconds(self): """ - A view to slice audio region by seconds (using ``region.seconds[start:end]``). + A view to slice audio region by seconds using + ``region.seconds[start:end]``. """ return self._seconds_view @property def millis(self): - """A view to slice audio region by milliseconds (using ``region.millis[start:end]``).""" + """A view to slice audio region by milliseconds using + ``region.millis[start:end]``.""" return self._millis_view @property @@ -786,7 +792,7 @@ max_silence=0.3, drop_trailing_silence=False, strict_min_dur=False, - **kwargs + **kwargs, ): """Split audio region. See :func:`auditok.split()` for a comprehensive description of split parameters. @@ -804,7 +810,7 @@ max_silence=max_silence, drop_trailing_silence=drop_trailing_silence, strict_min_dur=strict_min_dur, - **kwargs + **kwargs, ) def plot( @@ -816,7 +822,7 @@ dpi=120, theme="auditok", ): - """Plot audio region, one sub-plot for each channel. + """Plot audio region using one sub-plot per each channel. Parameters ---------- @@ -835,20 +841,15 @@ plot theme to use. Currently only "auditok" theme is implemented. To provide you own them see :attr:`auditok.plotting.AUDITOK_PLOT_THEME`. """ - try: - from auditok.plotting import plot - - plot( - self, - scale_signal=scale_signal, - show=show, - figsize=figsize, - save_as=save_as, - dpi=dpi, - theme=theme, - ) - except ImportError: - raise RuntimeWarning("Plotting requires matplotlib") + plot( + self, + scale_signal=scale_signal, + show=show, + figsize=figsize, + save_as=save_as, + dpi=dpi, + theme=theme, + ) def split_and_plot( self, @@ -863,42 +864,37 @@ save_as=None, dpi=120, theme="auditok", - **kwargs + **kwargs, ): """Split region and plot signal and detections. Alias: :meth:`splitp`. See :func:`auditok.split()` for a comprehensive description of split parameters. Also see :meth:`plot` for plot parameters. """ - try: - from auditok.plotting import plot - - regions = self.split( - min_dur=min_dur, - max_dur=max_dur, - max_silence=max_silence, - drop_trailing_silence=drop_trailing_silence, - strict_min_dur=strict_min_dur, - **kwargs - ) - regions = list(regions) - detections = ((reg.meta.start, reg.meta.end) for reg in regions) - eth = kwargs.get( - "energy_threshold", kwargs.get("eth", DEFAULT_ENERGY_THRESHOLD) - ) - plot( - self, - scale_signal=scale_signal, - detections=detections, - energy_threshold=eth, - show=show, - figsize=figsize, - save_as=save_as, - dpi=dpi, - theme=theme, - ) - return regions - except ImportError: - raise RuntimeWarning("Plotting requires matplotlib") + regions = self.split( + min_dur=min_dur, + max_dur=max_dur, + max_silence=max_silence, + drop_trailing_silence=drop_trailing_silence, + strict_min_dur=strict_min_dur, + **kwargs, + ) + regions = list(regions) + detections = ((reg.meta.start, reg.meta.end) for reg in regions) + eth = kwargs.get( + "energy_threshold", kwargs.get("eth", DEFAULT_ENERGY_THRESHOLD) + ) + plot( + self, + scale_signal=scale_signal, + detections=detections, + energy_threshold=eth, + show=show, + figsize=figsize, + save_as=save_as, + dpi=dpi, + theme=theme, + ) + return regions def __array__(self): return self.samples @@ -1116,13 +1112,11 @@ In that case the trailing silence can be removed if you use the `StreamTokenizer.DROP_TRAILING_SILENCE` mode. - -4 `(StreamTokenizer.STRICT_MIN_LENGTH | StreamTokenizer.DROP_TRAILING_SILENCE)`: + -4 `(StreamTokenizer.STRICT_MIN_LENGTH | StreamTokenizer.DROP_TRAILING_SILENCE)`: # noqa: B950 use both options. That means: first remove tailing silence, then check if the token still has a length of at least `min_length`. - - Examples --------
--- a/auditok/io.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/io.py Sun May 26 23:29:33 2024 +0200 @@ -15,12 +15,12 @@ to_file player_for """ + import os import sys import wave -import warnings from abc import ABC, abstractmethod -from functools import partial + from .exceptions import AudioIOError, AudioParameterError try: @@ -104,20 +104,17 @@ audio_parameters : tuple a tuple for audio parameters as (sampling_rate, sample_width, channels). """ - err_message = ( - "'{ln}' (or '{sn}') must be a positive integer, found: '{val}'" - ) parameters = [] - for (long_name, short_name) in ( + for long_name, short_name in ( ("sampling_rate", "sr"), ("sample_width", "sw"), ("channels", "ch"), ): param = param_dict.get(long_name, param_dict.get(short_name)) if param is None or not isinstance(param, int) or param <= 0: - raise AudioParameterError( - err_message.format(ln=long_name, sn=short_name, val=param) - ) + err_message = f"{long_name!r} (or {short_name!r}) must be a " + err_message += f"positive integer, passed value: {param}." + raise AudioParameterError(err_message) parameters.append(param) sampling_rate, sample_width, channels = parameters return sampling_rate, sample_width, channels @@ -141,7 +138,10 @@ """ def __init__( - self, sampling_rate, sample_width, channels, + self, + sampling_rate, + sample_width, + channels, ): if sample_width not in (1, 2, 4): @@ -283,9 +283,13 @@ """ def __init__( - self, data, sampling_rate=16000, sample_width=2, channels=1, + self, + data, + sampling_rate=16000, + sample_width=2, + channels=1, ): - AudioSource.__init__(self, sampling_rate, sample_width, channels) + super().__init__(sampling_rate, sample_width, channels) check_audio_data(data, sample_width, channels) self._data = data self._sample_size_all_channels = sample_width * channels @@ -558,7 +562,10 @@ """ def __init__( - self, sampling_rate=16000, sample_width=2, channels=1, + self, + sampling_rate=16000, + sample_width=2, + channels=1, ): FileAudioSource.__init__(self, sampling_rate, sample_width, channels) self._is_open = False @@ -610,7 +617,10 @@ """ def __init__( - self, sampling_rate=16000, sample_width=2, channels=1, + self, + sampling_rate=16000, + sample_width=2, + channels=1, ): if sample_width not in (1, 2, 4): raise ValueError("Sample width in bytes must be one of 1, 2 or 4") @@ -640,7 +650,7 @@ chunk_gen, total=nb_chunks, duration=duration, - **progress_bar_kwargs + **progress_bar_kwargs, ) if self.stream.is_stopped(): self.stream.start_stream() @@ -737,7 +747,7 @@ return PyAudioSource( *_get_audio_parameters(kwargs), frames_per_buffer=frames_per_buffer, - input_device_index=input_device_index + input_device_index=input_device_index, ) @@ -1004,12 +1014,7 @@ if audio_format in (None, "raw"): _save_raw(data, file) return - try: - sampling_rate, sample_width, channels = _get_audio_parameters(kwargs) - except AudioParameterError as exc: - err_message = "All audio parameters are required to save formats " - "other than raw. Error detail: {}".format(exc) - raise AudioParameterError(err_message) + sampling_rate, sample_width, channels = _get_audio_parameters(kwargs) if audio_format in ("wav", "wave"): _save_wave(data, file, sampling_rate, sample_width, channels) elif _WITH_PYDUB: @@ -1017,5 +1022,6 @@ data, file, audio_format, sampling_rate, sample_width, channels ) else: - err_message = "cannot write file format {} (file name: {})" - raise AudioIOError(err_message.format(audio_format, file)) + raise AudioIOError( + f"cannot write file format {audio_format} (file name: {file})" + )
--- a/auditok/plotting.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/plotting.py Sun May 26 23:29:33 2024 +0200 @@ -40,7 +40,7 @@ ls = theme.get("linestyle", theme.get("ls")) lw = theme.get("linewidth", theme.get("lw")) alpha = theme.get("alpha") - for (start, end) in detections: + for start, end in detections: subplot.axvspan(start, end, fc=fc, ec=ec, ls=ls, lw=lw, alpha=alpha)
--- a/auditok/signal.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/signal.py Sun May 26 23:29:33 2024 +0200 @@ -12,9 +12,10 @@ calculate_energy_single_channel calculate_energy_multichannel """ -from array import array as array_ + import audioop import math +from array import array as array_ FORMAT = {1: "b", 2: "h", 4: "i"} _EPSILON = 1e-10
--- a/auditok/signal_numpy.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/signal_numpy.py Sun May 26 23:29:33 2024 +0200 @@ -1,8 +1,9 @@ import numpy as np + from .signal import ( + calculate_energy_multichannel, + calculate_energy_single_channel, compute_average_channel_stereo, - calculate_energy_single_channel, - calculate_energy_multichannel, ) FORMAT = {1: np.int8, 2: np.int16, 4: np.int32}
--- a/auditok/workers.py Sun May 26 22:43:08 2024 +0200 +++ b/auditok/workers.py Sun May 26 23:29:33 2024 +0200 @@ -1,22 +1,22 @@ import os +import subprocess import sys +import wave +from abc import ABCMeta, abstractmethod +from collections import namedtuple +from datetime import datetime, timedelta +from queue import Empty, Queue from tempfile import NamedTemporaryFile -from abc import ABCMeta, abstractmethod from threading import Thread -from datetime import datetime, timedelta -from collections import namedtuple -import wave -import subprocess -from queue import Queue, Empty -from .io import _guess_audio_format -from .util import AudioDataSource, make_duration_formatter + from .core import split from .exceptions import ( - EndOfProcessing, AudioEncodingError, AudioEncodingWarning, + EndOfProcessing, ) - +from .io import _guess_audio_format +from .util import AudioReader, make_duration_formatter _STOP_PROCESSING = "STOP_PROCESSING" _Detection = namedtuple("_Detection", "id start end duration") @@ -86,7 +86,7 @@ return None -class TokenizerWorker(Worker, AudioDataSource): +class TokenizerWorker(Worker, AudioReader): def __init__(self, reader, observers=None, logger=None, **kwargs): self._observers = observers if observers is not None else [] self._reader = reader @@ -245,7 +245,7 @@ self.stop() def rewind(self): - # ensure compatibility with AudioDataSource with record=True + # ensure compatibility with AudioReader with record=True pass @property
--- a/doc/conf.py Sun May 26 22:43:08 2024 +0200 +++ b/doc/conf.py Sun May 26 23:29:33 2024 +0200 @@ -12,11 +12,11 @@ # All configuration values have a default; values that are commented out # serve to show the default. -import sys +import ast import os import re -import ast import shlex +import sys # If extensions (or modules to document with autodoc) are in another directory, # add these directories to sys.path here. If the directory is relative to the @@ -63,9 +63,9 @@ master_doc = "index" # General information about the project. -project = u"auditok" -copyright = u"2015-2021, Amine Sehili" -author = u"Amine Sehili" +project = "auditok" +copyright = "2015-2021, Amine Sehili" +author = "Amine Sehili" # The version info for the project you're documenting, acts as replacement for # |version| and |release|, also used in various other places throughout the @@ -253,8 +253,8 @@ ( master_doc, "auditok.tex", - u"auditok Documentation", - u"Amine Sehili", + "auditok Documentation", + "Amine Sehili", "manual", ), ] @@ -284,7 +284,7 @@ # One entry per manual page. List of tuples # (source start file, name, description, authors, manual section). -man_pages = [(master_doc, "auditok", u"auditok Documentation", [author], 1)] +man_pages = [(master_doc, "auditok", "auditok Documentation", [author], 1)] # If true, show URL addresses after external links. # man_show_urls = False @@ -299,7 +299,7 @@ ( master_doc, "auditok", - u"auditok Documentation", + "auditok Documentation", author, "auditok", "Audio Activity Detection tool.",