Skip to content

GitLab

Commit 95235f31 authored by jamarshon's avatar jamarshon Committed by cpuhrsch
Browse files

Large re-amp on the torchaudio/docs (#166)

parent 143f667f
Hide whitespace changes
Inline Side-by-side
Showing with 520 additions and 381 deletions
README.md
View file @ 95235f31
......@@ -11,7 +11,9 @@ torchaudio: an audio library for PyTorch
- [Kaldi (ark/scp)](http://pytorch.org/audio/kaldi_io.html)
- [Dataloaders for common audio datasets (VCTK, YesNo)](http://pytorch.org/audio/datasets.html)
- Common audio transforms
- [Scale, PadTrim, DownmixMono, LC2CL, BLC2CBL, MuLawEncoding, MuLawExpanding](http://pytorch.org/audio/transforms.html)
- [Spectrogram, SpectrogramToDB, MelScale, MelSpectrogram, MFCC, MuLawEncoding, MuLawDecoding, Resample](http://pytorch.org/audio/transforms.html)
- Compliance interfaces: Run code using PyTorch that align with other libraries
- [Kaldi: fbank, spectrogram, resample_waveform](https://pytorch.org/audio/compliance.kaldi.html)
Dependencies
------------
......
docs/source/compliance.kaldi.rst
View file @ 95235f31
......@@ -24,3 +24,8 @@ Functions
~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: spectrogram
:hidden:`resample_waveform`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: resample_waveform
docs/source/functional.rst
View file @ 95235f31
......@@ -8,63 +8,57 @@ torchaudio.functional
Functions to perform common audio operations.
:hidden:`scale`
~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: scale
:hidden:`pad_trim`
:hidden:`istft`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: pad_trim
.. autofunction:: istft
:hidden:`downmix_mono`
:hidden:`spectrogram`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: downmix_mono
.. autofunction:: spectrogram
:hidden:`LC2CL`
:hidden:`amplitude_to_DB`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: LC2CL
.. autofunction:: amplitude_to_DB
:hidden:`istft`
:hidden:`create_fb_matrix`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: istft
.. autofunction:: create_fb_matrix
:hidden:`spectrogram`
:hidden:`create_dct`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: spectrogram
.. autofunction:: create_dct
:hidden:`create_fb_matrix`
:hidden:`mu_law_encoding`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: create_fb_matrix
.. autofunction:: mu_law_encoding
:hidden:`spectrogram_to_DB`
:hidden:`mu_law_decoding`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: spectrogram_to_DB
.. autofunction:: mu_law_decoding
:hidden:`create_dct`
:hidden:`complex_norm`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: create_dct
.. autofunction:: complex_norm
:hidden:`BLC2CBL`
:hidden:`angle`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: BLC2CBL
.. autofunction:: angle
:hidden:`mu_law_encoding`
:hidden:`magphase`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: mu_law_encoding
.. autofunction:: magphase
:hidden:`mu_law_expanding`
:hidden:`phase_vocoder`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: mu_law_expanding
.. autofunction:: phase_vocoder
docs/source/kaldi_io.rst
View file @ 95235f31
......@@ -7,7 +7,7 @@ torchaudio.kaldi_io
.. currentmodule:: torchaudio.kaldi_io
To use this module, the dependency kaldi_io_ needs to be installed.
This is a light wrapper around ``kaldi_io`` that returns :class:`torch.Tensors`.
This is a light wrapper around ``kaldi_io`` that returns :class:`torch.Tensor`.
.. _kaldi_io: https://github.com/vesis84/kaldi-io-for-python
......
docs/source/legacy.rst
View file @ 95235f31
.. role:: hidden
:class: hidden-section
torchaudio.legacy
======================
.. currentmodule:: torchaudio.legacy
Legacy loading and save functions.
.. automodule:: torchaudio.legacy
:members:
:hidden:`load`
~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: load
:hidden:`save`
~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autofunction:: save
docs/source/sox_effects.rst
View file @ 95235f31
.. role:: hidden
:class: hidden-section
torchaudio.sox_effects
======================
......@@ -5,8 +8,14 @@ Create SoX effects chain for preprocessing audio.
.. currentmodule:: torchaudio.sox_effects
:hidden:`SoxEffect`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: SoxEffect
:members:
:hidden:`SoxEffectsChain`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: SoxEffectsChain
:members: append_effect_to_chain, sox_build_flow_effects, clear_chain, set_input_file
docs/source/transforms.rst
View file @ 95235f31
.. role:: hidden
:class: hidden-section
torchaudio.transforms
======================
.. currentmodule:: torchaudio.transforms
Transforms are common audio transforms. They can be chained together using :class:`Compose`
Transforms are common audio transforms. They can be chained together using :class:`torch.nn.Sequential`
:hidden:`Spectrogram`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: Spectrogram
.. automethod:: torchaudio._docs.Spectrogram.forward
:hidden:`AmplitudeToDB`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: AmplitudeToDB
.. automethod:: torchaudio._docs.AmplitudeToDB.forward
:hidden:`MelScale`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: Compose
.. autoclass:: MelScale
.. autoclass:: Scale
.. automethod:: torchaudio._docs.MelScale.forward
.. autoclass:: PadTrim
:hidden:`MelSpectrogram`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: DownmixMono
.. autoclass:: MelSpectrogram
.. autoclass:: LC2CL
.. automethod:: torchaudio._docs.MelSpectrogram.forward
.. autoclass:: MEL
:hidden:`MFCC`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: BLC2CBL
.. autoclass:: MFCC
.. automethod:: torchaudio._docs.MFCC.forward
:hidden:`MuLawEncoding`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: MuLawEncoding
.. autoclass:: MuLawExpanding
.. automethod:: torchaudio._docs.MuLawEncoding.forward
:hidden:`MuLawDecoding`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: MuLawDecoding
.. automethod:: torchaudio._docs.MuLawDecoding.forward
:hidden:`Resample`
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
.. autoclass:: Resample
.. automethod:: torchaudio._docs.Resample.forward
torchaudio/__init__.py
View file @ 95235f31
......@@ -5,7 +5,7 @@ import torch
import _torch_sox
from .version import __version__, git_version
from torchaudio import transforms, datasets, kaldi_io, sox_effects, legacy, compliance
from torchaudio import transforms, datasets, kaldi_io, sox_effects, legacy, compliance, _docs
def check_input(src):
......@@ -24,33 +24,35 @@ def load(filepath,
signalinfo=None,
encodinginfo=None,
filetype=None):
"""Loads an audio file from disk into a Tensor
r"""Loads an audio file from disk into a tensor
Args:
filepath (string or pathlib.Path): path to audio file
out (Tensor, optional): an output Tensor to use instead of creating one
filepath (str or pathlib.Path): Path to audio file
out (torch.Tensor, optional): An output tensor to use instead of creating one. (Default: ``None``)
normalization (bool, number, or callable, optional): If boolean `True`, then output is divided by `1 << 31`
(assumes signed 32-bit audio), and normalizes to `[0, 1]`.
If `number`, then output is divided by that number
If `callable`, then the output is passed as a parameter
to the given function, then the output is divided by
the result.
channels_first (bool): Set channels first or length first in result. Default: ``True``
num_frames (int, optional): number of frames to load. 0 to load everything after the offset.
offset (int, optional): number of frames from the start of the file to begin data loading.
signalinfo (sox_signalinfo_t, optional): a sox_signalinfo_t type, which could be helpful if the
audio type cannot be automatically determined
encodinginfo (sox_encodinginfo_t, optional): a sox_encodinginfo_t type, which could be set if the
audio type cannot be automatically determined
filetype (str, optional): a filetype or extension to be set if sox cannot determine it automatically
Returns: tuple(Tensor, int)
- Tensor: output Tensor of size `[C x L]` or `[L x C]` where L is the number of audio frames and
C is the number of channels
- int: the sample rate of the audio (as listed in the metadata of the file)
Example::
(assumes signed 32-bit audio), and normalizes to `[0, 1]`.
If `number`, then output is divided by that number
If `callable`, then the output is passed as a parameter
to the given function, then the output is divided by
the result. (Default: ``True``)
channels_first (bool): Set channels first or length first in result. (Default: ``True``)
num_frames (int, optional): Number of frames to load. 0 to load everything after the offset.
(Default: ``0``)
offset (int, optional): Number of frames from the start of the file to begin data loading.
(Default: ``0``)
signalinfo (sox_signalinfo_t, optional): A sox_signalinfo_t type, which could be helpful if the
audio type cannot be automatically determined. (Default: ``None``)
encodinginfo (sox_encodinginfo_t, optional): A sox_encodinginfo_t type, which could be set if the
audio type cannot be automatically determined. (Default: ``None``)
filetype (str, optional): A filetype or extension to be set if sox cannot determine it
automatically. (Default: ``None``)
Returns:
Tuple[torch.Tensor, int]: An output tensor of size `[C x L]` or `[L x C]` where L is the number
of audio frames and C is the number of channels. An integer which is the sample rate of the
audio (as listed in the metadata of the file)
Example
>>> data, sample_rate = torchaudio.load('foo.mp3')
>>> print(data.size())
torch.Size([2, 278756])
......@@ -94,16 +96,33 @@ def load(filepath,
def load_wav(filepath, **kwargs):
""" Loads a wave file. It assumes that the wav file uses 16 bit per sample that needs normalization by shifting
r""" Loads a wave file. It assumes that the wav file uses 16 bit per sample that needs normalization by shifting
the input right by 16 bits.
Args:
filepath (str or pathlib.Path): Path to audio file
Returns:
Tuple[torch.Tensor, int]: An output tensor of size `[C x L]` or `[L x C]` where L is the number
of audio frames and C is the number of channels. An integer which is the sample rate of the
audio (as listed in the metadata of the file)
"""
kwargs['normalization'] = 1 << 16
return load(filepath, **kwargs)
def save(filepath, src, sample_rate, precision=16, channels_first=True):
"""Convenience function for `save_encinfo`.
r"""Convenience function for `save_encinfo`.
Args:
filepath (str): Path to audio file
src (torch.Tensor): An input 2D tensor of shape `[C x L]` or `[L x C]` where L is
the number of audio frames, C is the number of channels
sample_rate (int): An integer which is the sample rate of the
audio (as listed in the metadata of the file)
precision (int): Bit precision (Default: ``16``)
channels_first (bool): Set channels first or length first in result. (
Default: ``True``)
"""
si = sox_signalinfo_t()
ch_idx = 0 if channels_first else 1
......@@ -120,21 +139,21 @@ def save_encinfo(filepath,
signalinfo=None,
encodinginfo=None,
filetype=None):
"""Saves a Tensor of an audio signal to disk as a standard format like mp3, wav, etc.
r"""Saves a tensor of an audio signal to disk as a standard format like mp3, wav, etc.
Args:
filepath (string): path to audio file
src (Tensor): an input 2D Tensor of shape `[C x L]` or `[L x C]` where L is
the number of audio frames, C is the number of channels
channels_first (bool): Set channels first or length first in result. Default: ``True``
signalinfo (sox_signalinfo_t): a sox_signalinfo_t type, which could be helpful if the
audio type cannot be automatically determined
encodinginfo (sox_encodinginfo_t, optional): a sox_encodinginfo_t type, which could be set if the
audio type cannot be automatically determined
filetype (str, optional): a filetype or extension to be set if sox cannot determine it automatically
Example::
filepath (str): Path to audio file
src (torch.Tensor): An input 2D tensor of shape `[C x L]` or `[L x C]` where L is
the number of audio frames, C is the number of channels
channels_first (bool): Set channels first or length first in result. (Default: ``True``)
signalinfo (sox_signalinfo_t): A sox_signalinfo_t type, which could be helpful if the
audio type cannot be automatically determined. (Default: ``None``)
encodinginfo (sox_encodinginfo_t, optional): A sox_encodinginfo_t type, which could be set if the
audio type cannot be automatically determined. (Default: ``None``)
filetype (str, optional): A filetype or extension to be set if sox cannot determine it
automatically. (Default: ``None``)
Example
>>> data, sample_rate = torchaudio.load('foo.mp3')
>>> torchaudio.save('foo.wav', data, sample_rate)
......@@ -184,16 +203,16 @@ def save_encinfo(filepath,
def info(filepath):
"""Gets metadata from an audio file without loading the signal.
r"""Gets metadata from an audio file without loading the signal.
Args:
filepath (string): path to audio file
filepath (str): Path to audio file
Returns: tuple(si, ei)
- si (sox_signalinfo_t): signal info as a python object
- ei (sox_encodinginfo_t): encoding info as a python object
Returns:
Tuple[sox_signalinfo_t, sox_encodinginfo_t]: A si (sox_signalinfo_t) signal
info as a python object. An ei (sox_encodinginfo_t) encoding info
Example::
Example
>>> si, ei = torchaudio.info('foo.wav')
>>> rate, channels, encoding = si.rate, si.channels, ei.encoding
"""
......@@ -206,13 +225,13 @@ def sox_signalinfo_t():
primarily for effects
Returns: sox_signalinfo_t(object)
- rate (float), sample rate as a float, practically will likely be an integer float
- channel (int), number of audio channels
- precision (int), bit precision
- length (int), length of audio in samples * channels, 0 for unspecified and -1 for unknown
- mult (float, optional), headroom multiplier for effects and None for no multiplier
- rate (float), sample rate as a float, practically will likely be an integer float
- channel (int), number of audio channels
- precision (int), bit precision
- length (int), length of audio in samples * channels, 0 for unspecified and -1 for unknown
- mult (float, optional), headroom multiplier for effects and ``None`` for no multiplier
Example::
Example
>>> si = torchaudio.sox_signalinfo_t()
>>> si.channels = 1
>>> si.rate = 16000.
......@@ -223,7 +242,7 @@ def sox_signalinfo_t():
def sox_encodinginfo_t():
"""Create a sox_encodinginfo_t object. This object can be used to set the encoding
r"""Create a sox_encodinginfo_t object. This object can be used to set the encoding
type, bit precision, compression factor, reverse bytes, reverse nibbles,
reverse bits and endianness. This can be used in an effects chain to encode the
final output or to save a file with a specific encoding. For example, one could
......@@ -232,15 +251,15 @@ def sox_encodinginfo_t():
the bit precision.
Returns: sox_encodinginfo_t(object)
- encoding (sox_encoding_t), output encoding
- bits_per_sample (int), bit precision, same as `precision` in sox_signalinfo_t
- compression (float), compression for lossy formats, 0.0 for default compression
- reverse_bytes (sox_option_t), reverse bytes, use sox_option_default
- reverse_nibbles (sox_option_t), reverse nibbles, use sox_option_default
- reverse_bits (sox_option_t), reverse bytes, use sox_option_default
- opposite_endian (sox_bool), change endianness, use sox_false
Example::
- encoding (sox_encoding_t), output encoding
- bits_per_sample (int), bit precision, same as `precision` in sox_signalinfo_t
- compression (float), compression for lossy formats, 0.0 for default compression
- reverse_bytes (sox_option_t), reverse bytes, use sox_option_default
- reverse_nibbles (sox_option_t), reverse nibbles, use sox_option_default
- reverse_bits (sox_option_t), reverse bytes, use sox_option_default
- opposite_endian (sox_bool), change endianness, use sox_false
Example
>>> ei = torchaudio.sox_encodinginfo_t()
>>> ei.encoding = torchaudio.get_sox_encoding_t(1)
>>> ei.bits_per_sample = 16
......@@ -260,13 +279,14 @@ def sox_encodinginfo_t():
def get_sox_encoding_t(i=None):
"""Get enum of sox_encoding_t for sox encodings.
r"""Get enum of sox_encoding_t for sox encodings.
Args:
i (int, optional): choose type or get a dict with all possible options
use `__members__` to see all options when not specified
i (int, optional): Choose type or get a dict with all possible options
use ``__members__`` to see all options when not specified. (Default: ``None``)
Returns:
sox_encoding_t: a sox_encoding_t type for output encoding
sox_encoding_t: A sox_encoding_t type for output encoding
"""
if i is None:
# one can see all possible values using the .__members__ attribute
......@@ -276,14 +296,14 @@ def get_sox_encoding_t(i=None):
def get_sox_option_t(i=2):
"""Get enum of sox_option_t for sox encodinginfo options.
r"""Get enum of sox_option_t for sox encodinginfo options.
Args:
i (int, optional): choose type or get a dict with all possible options
use `__members__` to see all options when not specified.
Defaults to sox_option_default.
i (int, optional): Choose type or get a dict with all possible options
use ``__members__`` to see all options when not specified.
(Default: ``sox_option_default`` or ``2``)
Returns:
sox_option_t: a sox_option_t type
sox_option_t: A sox_option_t type
"""
if i is None:
return _torch_sox.sox_option_t
......@@ -292,14 +312,15 @@ def get_sox_option_t(i=2):
def get_sox_bool(i=0):
"""Get enum of sox_bool for sox encodinginfo options.
r"""Get enum of sox_bool for sox encodinginfo options.
Args:
i (int, optional): choose type or get a dict with all possible options
use `__members__` to see all options when not specified.
Defaults to sox_false.
i (int, optional): Choose type or get a dict with all possible options
use ``__members__`` to see all options when not specified. (Default:
``sox_false`` or ``0``)
Returns:
sox_bool: a sox_bool type
sox_bool: A sox_bool type
"""
if i is None:
return _torch_sox.sox_bool
......
torchaudio/_docs.py 0 → 100644
View file @ 95235f31
import torchaudio
# TODO See https://github.com/pytorch/audio/issues/165
class Spectrogram:
forward = torchaudio.transforms.Spectrogram().forward
class AmplitudeToDB:
forward = torchaudio.transforms.AmplitudeToDB().forward
class MelScale:
forward = torchaudio.transforms.MelScale().forward
class MelSpectrogram:
forward = torchaudio.transforms.MelSpectrogram().forward
class MFCC:
forward = torchaudio.transforms.MFCC().forward
class MuLawEncoding:
forward = torchaudio.transforms.MuLawEncoding().forward
class MuLawDecoding:
forward = torchaudio.transforms.MuLawDecoding().forward
class Resample:
# Resample isn't a script_method
forward = torchaudio.transforms.Resample.forward
torchaudio/compliance/kaldi.py
View file @ 95235f31
......@@ -37,11 +37,11 @@ def _next_power_of_2(x):
def _get_strided(waveform, window_size, window_shift, snip_edges):
r"""Given a waveform (1D tensor of size `num_samples`), it returns a 2D tensor (m, `window_size`)
r"""Given a waveform (1D tensor of size ``num_samples``), it returns a 2D tensor (m, ``window_size``)
representing how the window is shifted along the waveform. Each row is a frame.
Args:
waveform (torch.Tensor): Tensor of size `num_samples`
waveform (torch.Tensor): Tensor of size ``num_samples``
window_size (int): Frame length
window_shift (int): Frame shift
snip_edges (bool): If True, end effects will be handled by outputting only frames that completely fit
......@@ -49,7 +49,7 @@ def _get_strided(waveform, window_size, window_shift, snip_edges):
depends only on the frame_shift, and we reflect the data at the ends.
Returns:
torch.Tensor: 2D tensor of size (m, `window_size`) where each row is a frame
torch.Tensor: 2D tensor of size (m, ``window_size``) where each row is a frame
"""
assert waveform.dim() == 1
num_samples = waveform.size(0)
......@@ -134,7 +134,7 @@ def _get_window(waveform, padded_window_size, window_size, window_shift, window_
r"""Gets a window and its log energy
Returns:
strided_input (torch.Tensor): size (m, `padded_window_size`)
strided_input (torch.Tensor): size (m, ``padded_window_size``)
signal_log_energy (torch.Tensor): size (m)
"""
# size (m, window_size)
......@@ -191,33 +191,33 @@ def spectrogram(
Args:
waveform (torch.Tensor): Tensor of audio of size (c, n) where c is in the range [0,2)
blackman_coeff (float): Constant coefficient for generalized Blackman window. (Default: 0.42)
channel (int): Channel to extract (-1 -> expect mono, 0 -> left, 1 -> right) (Default: -1)
blackman_coeff (float): Constant coefficient for generalized Blackman window. (Default: ``0.42``)
channel (int): Channel to extract (-1 -> expect mono, 0 -> left, 1 -> right) (Default: ``-1``)
dither (float): Dithering constant (0.0 means no dither). If you turn this off, you should set
the energy_floor option, e.g. to 1.0 or 0.1 (Default: 1.0)
the energy_floor option, e.g. to 1.0 or 0.1 (Default: ``1.0``)
energy_floor (float): Floor on energy (absolute, not relative) in Spectrogram computation. Caution:
this floor is applied to the zeroth component, representing the total signal energy. The floor on the
individual spectrogram elements is fixed at std::numeric_limits<float>::epsilon(). (Default: 0.0)
frame_length (float): Frame length in milliseconds (Default: 25.0)
frame_shift (float): Frame shift in milliseconds (Default: 10.0)
min_duration (float): Minimum duration of segments to process (in seconds). (Default: 0.0)
preemphasis_coefficient (float): Coefficient for use in signal preemphasis (Default: 0.97)
raw_energy (bool): If True, compute energy before preemphasis and windowing (Default: True)
remove_dc_offset: Subtract mean from waveform on each frame (Default: True)
individual spectrogram elements is fixed at std::numeric_limits<float>::epsilon(). (Default: ``0.0``)
frame_length (float): Frame length in milliseconds (Default: ``25.0``)
frame_shift (float): Frame shift in milliseconds (Default: ``10.0``)
min_duration (float): Minimum duration of segments to process (in seconds). (Default: ``0.0``)
preemphasis_coefficient (float): Coefficient for use in signal preemphasis (Default: ``0.97``)
raw_energy (bool): If True, compute energy before preemphasis and windowing (Default: ``True``)
remove_dc_offset: Subtract mean from waveform on each frame (Default: ``True``)
round_to_power_of_two (bool): If True, round window size to power of two by zero-padding input
to FFT. (Default: True)
to FFT. (Default: ``True``)
sample_frequency (float): Waveform data sample frequency (must match the waveform file, if
specified there) (Default: 16000.0)
specified there) (Default: ``16000.0``)
snip_edges (bool): If True, end effects will be handled by outputting only frames that completely fit
in the file, and the number of frames depends on the frame_length. If False, the number of frames
depends only on the frame_shift, and we reflect the data at the ends. (Default: True)
depends only on the frame_shift, and we reflect the data at the ends. (Default: ``True``)
subtract_mean (bool): Subtract mean of each feature file [CMS]; not recommended to do
it this way. (Default: False)
window_type (str): Type of window ('hamming'|'hanning'|'povey'|'rectangular'|'blackman') (Default: 'povey')
it this way. (Default: ``False``)
window_type (str): Type of window ('hamming'|'hanning'|'povey'|'rectangular'|'blackman') (Default: ``'povey'``)
Returns:
torch.Tensor: A spectrogram identical to what Kaldi would output. The shape is
(m, `padded_window_size` // 2 + 1) where m is calculated in _get_strided
(m, ``padded_window_size // 2 + 1``) where m is calculated in _get_strided
"""
waveform, window_shift, window_size, padded_window_size = _get_waveform_and_window_properties(
waveform, channel, sample_frequency, frame_shift, frame_length, round_to_power_of_two, preemphasis_coefficient)
......@@ -343,7 +343,7 @@ def vtln_warp_mel_freq(vtln_low_cutoff, vtln_high_cutoff, low_freq, high_freq,
mel_freq (torch.Tensor): Given frequency in Mel
Returns:
torch.Tensor: `mel_freq` after vtln warp
torch.Tensor: ``mel_freq`` after vtln warp
"""
return mel_scale(vtln_warp_freq(vtln_low_cutoff, vtln_high_cutoff, low_freq, high_freq,
vtln_warp_factor, inverse_mel_scale(mel_freq)))
......@@ -354,9 +354,9 @@ def get_mel_banks(num_bins, window_length_padded, sample_freq,
# type: (int, int, float, float, float, float, float)
"""
Returns:
Tuple[torch.Tensor, torch.Tensor]: The tuple consists of `bins` (which is
Melbank of size (`num_bins`, `num_fft_bins`)) and `center_freqs` (which is
Center frequencies of bins of size (`num_bins`)).
Tuple[torch.Tensor, torch.Tensor]: The tuple consists of ``bins`` (which is
melbank of size (``num_bins``, ``num_fft_bins``)) and ``center_freqs`` (which is
center frequencies of bins of size (``num_bins``)).
"""
assert num_bins > 3, 'Must have at least 3 mel bins'
assert window_length_padded % 2 == 0
......@@ -430,44 +430,44 @@ def fbank(
Args:
waveform (torch.Tensor): Tensor of audio of size (c, n) where c is in the range [0,2)
blackman_coeff (float): Constant coefficient for generalized Blackman window. (Default: 0.42)
channel (int): Channel to extract (-1 -> expect mono, 0 -> left, 1 -> right) (Default: -1)
blackman_coeff (float): Constant coefficient for generalized Blackman window. (Default: ``0.42``)
channel (int): Channel to extract (-1 -> expect mono, 0 -> left, 1 -> right) (Default: ``-1``)
dither (float): Dithering constant (0.0 means no dither). If you turn this off, you should set
the energy_floor option, e.g. to 1.0 or 0.1 (Default: 1.0)
the energy_floor option, e.g. to 1.0 or 0.1 (Default: ``1.0``)
energy_floor (float): Floor on energy (absolute, not relative) in Spectrogram computation. Caution:
this floor is applied to the zeroth component, representing the total signal energy. The floor on the
individual spectrogram elements is fixed at std::numeric_limits<float>::epsilon(). (Default: 0.0)
frame_length (float): Frame length in milliseconds (Default: 25.0)
frame_shift (float): Frame shift in milliseconds (Default: 10.0)
high_freq (float): High cutoff frequency for mel bins (if <= 0, offset from Nyquist) (Default: 0.0)
individual spectrogram elements is fixed at std::numeric_limits<float>::epsilon(). (Default: ``0.0``)
frame_length (float): Frame length in milliseconds (Default: ``25.0``)
frame_shift (float): Frame shift in milliseconds (Default: ``10.0``)
high_freq (float): High cutoff frequency for mel bins (if <= 0, offset from Nyquist) (Default: ``0.0``)
htk_compat (bool): If true, put energy last. Warning: not sufficient to get HTK compatible features (need
to change other parameters). (Default: False)
low_freq (float): Low cutoff frequency for mel bins (Default: 20.0)
min_duration (float): Minimum duration of segments to process (in seconds). (Default: 0.0)
num_mel_bins (int): Number of triangular mel-frequency bins (Default: 23)
preemphasis_coefficient (float): Coefficient for use in signal preemphasis (Default: 0.97)
raw_energy (bool): If True, compute energy before preemphasis and windowing (Default: True)
remove_dc_offset: Subtract mean from waveform on each frame (Default: True)
to change other parameters). (Default: ``False``)
low_freq (float): Low cutoff frequency for mel bins (Default: ``20.0``)
min_duration (float): Minimum duration of segments to process (in seconds). (Default: ``0.0``)
num_mel_bins (int): Number of triangular mel-frequency bins (Default: ``23``)
preemphasis_coefficient (float): Coefficient for use in signal preemphasis (Default: ``0.97``)
raw_energy (bool): If True, compute energy before preemphasis and windowing (Default: ``True``)
remove_dc_offset: Subtract mean from waveform on each frame (Default: ``True``)
round_to_power_of_two (bool): If True, round window size to power of two by zero-padding input
to FFT. (Default: True)
to FFT. (Default: ``True``)
sample_frequency (float): Waveform data sample frequency (must match the waveform file, if
specified there) (Default: 16000.0)
specified there) (Default: ``16000.0``)
snip_edges (bool): If True, end effects will be handled by outputting only frames that completely fit
in the file, and the number of frames depends on the frame_length. If False, the number of frames
depends only on the frame_shift, and we reflect the data at the ends. (Default: True)
depends only on the frame_shift, and we reflect the data at the ends. (Default: ``True``)
subtract_mean (bool): Subtract mean of each feature file [CMS]; not recommended to do
it this way. (Default: False)
use_energy (bool): Add an extra dimension with energy to the FBANK output. (Default: False)
use_log_fbank (bool):If true, produce log-filterbank, else produce linear. (Default: True)
use_power (bool): If true, use power, else use magnitude. (Default: True)
it this way. (Default: ``False``)
use_energy (bool): Add an extra dimension with energy to the FBANK output. (Default: ``False``)
use_log_fbank (bool):If true, produce log-filterbank, else produce linear. (Default: ``True``)
use_power (bool): If true, use power, else use magnitude. (Default: ``True``)
vtln_high (float): High inflection point in piecewise linear VTLN warping function (if
negative, offset from high-mel-freq (Default: -500.0)
vtln_low (float): Low inflection point in piecewise linear VTLN warping function (Default: 100.0)
vtln_warp (float): Vtln warp factor (only applicable if vtln_map not specified) (Default: 1.0)
window_type (str): Type of window ('hamming'|'hanning'|'povey'|'rectangular'|'blackman') (Default: 'povey')
negative, offset from high-mel-freq (Default: ``-500.0``)
vtln_low (float): Low inflection point in piecewise linear VTLN warping function (Default: ``100.0``)
vtln_warp (float): Vtln warp factor (only applicable if vtln_map not specified) (Default: ``1.0``)
window_type (str): Type of window ('hamming'|'hanning'|'povey'|'rectangular'|'blackman') (Default: ``'povey'``)
Returns:
torch.Tensor: A fbank identical to what Kaldi would output. The shape is (m, `num_mel_bins` + `use_energy`)
torch.Tensor: A fbank identical to what Kaldi would output. The shape is (m, ``num_mel_bins + use_energy``)
where m is calculated in _get_strided
"""
waveform, window_shift, window_size, padded_window_size = _get_waveform_and_window_properties(
......@@ -523,7 +523,7 @@ def _get_LR_indices_and_weights(orig_freq, new_freq, output_samples_in_unit, win
r"""Based on LinearResample::SetIndexesAndWeights where it retrieves the weights for
resampling as well as the indices in which they are valid. LinearResample (LR) means
that the output signal is at linearly spaced intervals (i.e the output signal has a
frequency of `new_freq`). It uses sinc/bandlimited interpolation to upsample/downsample
frequency of ``new_freq``). It uses sinc/bandlimited interpolation to upsample/downsample
the signal.
The reason why the same filter is not used for multiple convolutions is because the
......@@ -541,7 +541,7 @@ def _get_LR_indices_and_weights(orig_freq, new_freq, output_samples_in_unit, win
assuming the center of the sinc function is at 0, 16, and 32 (the deltas [..., 6, 1, 4, ....]
for 16 vs [...., 2, 3, ....] for 32)
Example, one case is when the orig_freq and new_freq are multiples of each other then
Example, one case is when the ``orig_freq`` and ``new_freq`` are multiples of each other then
there needs to be one filter.
A windowed filter function (i.e. Hanning * sinc) because the ideal case of sinc function
......@@ -562,9 +562,9 @@ def _get_LR_indices_and_weights(orig_freq, new_freq, output_samples_in_unit, win
efficient. We suggest around 4 to 10 for normal use
Returns:
Tuple[torch.Tensor, torch.Tensor]: A tuple of `min_input_index` (which is the minimum indices
where the window is valid, size (`output_samples_in_unit`)) and `weights` (which is the weights
which correspond with min_input_index, size (`output_samples_in_unit`, `max_weight_width`)).
Tuple[torch.Tensor, torch.Tensor]: A tuple of ``min_input_index`` (which is the minimum indices
where the window is valid, size (``output_samples_in_unit``)) and ``weights`` (which is the weights
which correspond with min_input_index, size (``output_samples_in_unit``, ``max_weight_width``)).
"""
assert lowpass_cutoff < min(orig_freq, new_freq) / 2
output_t = torch.arange(0, output_samples_in_unit, dtype=torch.get_default_dtype()) / new_freq
......@@ -606,7 +606,7 @@ def _lcm(a, b):
def _get_num_LR_output_samples(input_num_samp, samp_rate_in, samp_rate_out):
r"""Based on LinearResample::GetNumOutputSamples. LinearResample (LR) means that
the output signal is at linearly spaced intervals (i.e the output signal has a
frequency of `new_freq`). It uses sinc/bandlimited interpolation to upsample/downsample
frequency of ``new_freq``). It uses sinc/bandlimited interpolation to upsample/downsample
the signal.
Args:
......@@ -651,7 +651,7 @@ def resample_waveform(waveform, orig_freq, new_freq, lowpass_filter_width=6):
r"""Resamples the waveform at the new frequency. This matches Kaldi's OfflineFeatureTpl ResampleWaveform
which uses a LinearResample (resample a signal at linearly spaced intervals to upsample/downsample
a signal). LinearResample (LR) means that the output signal is at linearly spaced intervals (i.e
the output signal has a frequency of `new_freq`). It uses sinc/bandlimited interpolation to
the output signal has a frequency of ``new_freq``). It uses sinc/bandlimited interpolation to
upsample/downsample the signal.
https://ccrma.stanford.edu/~jos/resample/Theory_Ideal_Bandlimited_Interpolation.html
......@@ -662,10 +662,10 @@ def resample_waveform(waveform, orig_freq, new_freq, lowpass_filter_width=6):
orig_freq (float): The original frequency of the signal
new_freq (float): The desired frequency
lowpass_filter_width (int): Controls the sharpness of the filter, more == sharper
but less efficient. We suggest around 4 to 10 for normal use. (Default: 6)
but less efficient. We suggest around 4 to 10 for normal use. (Default: ``6``)
Returns:
torch.Tensor: The signal at the new frequency
torch.Tensor: The waveform at the new frequency
"""
assert waveform.dim() == 2
assert orig_freq > 0.0 and new_freq > 0.0
......
torchaudio/datasets/vctk.py
View file @ 95235f31
......@@ -71,21 +71,22 @@ def load_txts(dir):
class VCTK(data.Dataset):
"""`VCTK <http://homepages.inf.ed.ac.uk/jyamagis/page3/page58/page58.html>`_ Dataset.
`alternate url <http://datashare.is.ed.ac.uk/handle/10283/2651>`
r"""`VCTK <http://homepages.inf.ed.ac.uk/jyamagis/page3/page58/page58.html>`_ Dataset.
`alternate url <http://datashare.is.ed.ac.uk/handle/10283/2651>`_
Args:
root (string): Root directory of dataset where ``processed/training.pt``
root (str): Root directory of dataset where ``processed/training.pt``
and ``processed/test.pt`` exist.
downsample (bool, optional): Whether to downsample the signal (Default: ``True``)
transform (Callable, optional): A function/transform that takes in an raw audio
and returns a transformed version. E.g, ``transforms.Spectrogram``. (Default: ``None``)
target_transform (callable, optional): A function/transform that takes in the
target and transforms it. (Default: ``None``)
download (bool, optional): If true, downloads the dataset from the internet and
puts it in root directory. If dataset is already downloaded, it is not
downloaded again.
transform (callable, optional): A function/transform that takes in an raw audio
and returns a transformed version. E.g, ``transforms.Scale``
target_transform (callable, optional): A function/transform that takes in the
target and transforms it.
dev_mode(bool, optional): if true, clean up is not performed on downloaded
files. Useful to keep raw audio and transcriptions.
downloaded again. (Default: ``True``)
dev_mode(bool, optional): If true, clean up is not performed on downloaded
files. Useful to keep raw audio and transcriptions. (Default: ``False``)
"""
raw_folder = 'vctk/raw'
processed_folder = 'vctk/processed'
......@@ -121,7 +122,8 @@ class VCTK(data.Dataset):
index (int): Index
Returns:
tuple: (image, target) where target is index of the target class.
Tuple[torch.Tensor, int]: The output tuple (image, target) where target
is index of the target class.
"""
if self.cached_pt != index // self.chunk_size:
self.cached_pt = int(index // self.chunk_size)
......
torchaudio/datasets/yesno.py
View file @ 95235f31
......@@ -9,20 +9,21 @@ import torchaudio
class YESNO(data.Dataset):
"""`YesNo Hebrew <http://www.openslr.org/1/>`_ Dataset.
r"""`YesNo Hebrew <http://www.openslr.org/1/>`_ Dataset.
Args:
root (string): Root directory of dataset where ``processed/training.pt``
root (str): Root directory of dataset where ``processed/training.pt``
and ``processed/test.pt`` exist.
transform (Callable, optional): A function/transform that takes in an PIL image
and returns a transformed version. E.g, ``transforms.Spectrogram``. (
Default: ``None``)
target_transform (Callable, optional): A function/transform that takes in the
target and transforms it. (Default: ``None``)
download (bool, optional): If true, downloads the dataset from the internet and
puts it in root directory. If dataset is already downloaded, it is not
downloaded again.
transform (callable, optional): A function/transform that takes in an PIL image
and returns a transformed version. E.g, ``transforms.Scale``
target_transform (callable, optional): A function/transform that takes in the
target and transforms it.
dev_mode(bool, optional): if true, clean up is not performed on downloaded
files. Useful to keep raw audio and transcriptions.
downloaded again. (Default: ``False``)
dev_mode(bool, optional): If true, clean up is not performed on downloaded
files. Useful to keep raw audio and transcriptions. (Default: ``False``)
"""
raw_folder = 'yesno/raw'
processed_folder = 'yesno/processed'
......@@ -55,7 +56,8 @@ class YESNO(data.Dataset):
index (int): Index
Returns:
tuple: (image, target) where target is index of the target class.
Tuple[torch.Tensor, int]: The output tuple (image, target) where target
is index of the target class.
"""
audio, target = self.data[index], self.labels[index]
......
torchaudio/functional.py
View file @ 95235f31
......@@ -36,7 +36,7 @@ def istft(stft_matrix, # type: Tensor
length=None # type: Optional[int]
):
# type: (...) -> Tensor
r""" Inverse short time Fourier Transform. This is expected to be the inverse of torch.stft.
r"""Inverse short time Fourier Transform. This is expected to be the inverse of torch.stft.
It has the same parameters (+ additional optional parameter of ``length``) and it should return the
least squares estimation of the original signal. The algorithm will check using the NOLA condition (
nonzero overlap).
......@@ -46,7 +46,7 @@ def istft(stft_matrix, # type: Tensor
:math:`\sum_{t=-\infty}^{\infty} w^2[n-t\times hop\_length] \cancel{=} 0`.
Since stft discards elements at the end of the signal if they do not fit in a frame, the
istft may return a shorter signal than the original signal (can occur if `center` is False
istft may return a shorter signal than the original signal (can occur if ``center`` is False
since the signal isn't padded).
If ``center`` is True, then there will be padding e.g. 'constant', 'reflect', etc. Left padding
......@@ -66,7 +66,7 @@ def istft(stft_matrix, # type: Tensor
Args:
stft_matrix (torch.Tensor): Output of stft where each row of a channel is a frequency and each
column is a window. it has a shape of either (channel, fft_size, n_frames, 2) or (
column is a window. it has a size of either (channel, fft_size, n_frames, 2) or (
fft_size, n_frames, 2)
n_fft (int): Size of Fourier transform
hop_length (Optional[int]): The distance between neighboring sliding window frames.
......@@ -75,10 +75,12 @@ def istft(stft_matrix, # type: Tensor
window (Optional[torch.Tensor]): The optional window function.
(Default: ``torch.ones(win_length)``)
center (bool): Whether ``input`` was padded on both sides so
that the :math:`t`-th frame is centered at time :math:`t \times \text{hop\_length}`
pad_mode (str): Controls the padding method used when ``center`` is ``True``
normalized (bool): Whether the STFT was normalized
onesided (bool): Whether the STFT is onesided
that the :math:`t`-th frame is centered at time :math:`t \times \text{hop\_length}`.
(Default: ``True``)
pad_mode (str): Controls the padding method used when ``center`` is True. (Default:
``'reflect'``)
normalized (bool): Whether the STFT was normalized. (Default: ``False``)
onesided (bool): Whether the STFT is onesided. (Default: ``True``)
length (Optional[int]): The amount to trim the signal by (i.e. the
original signal length). (Default: whole signal)
......@@ -175,10 +177,10 @@ def spectrogram(waveform, pad, window, n_fft, hop_length, win_length, power, nor
r"""Create a spectrogram from a raw audio signal.
Args:
waveform (torch.Tensor): Tensor of audio of size (c, n)
waveform (torch.Tensor): Tensor of audio of dimension (channel, time)
pad (int): Two sided padding of signal
window (torch.Tensor): Window tensor that is applied/multiplied to each frame/window
n_fft (int): Size of fft
n_fft (int): Size of FFT
hop_length (int): Length of hop between STFT windows
win_length (int): Window size
power (int): Exponent for the magnitude spectrogram,
......@@ -186,9 +188,9 @@ def spectrogram(waveform, pad, window, n_fft, hop_length, win_length, power, nor
normalized (bool): Whether to normalize by magnitude after stft
Returns:
torch.Tensor: Channels x frequency x time (c, f, t), where channels
is unchanged, frequency is `n_fft // 2 + 1` where `n_fft` is the number of
fourier bins, and time is the number of window hops (n_frames).
torch.Tensor: Dimension (channel, freq, time), where channel
is unchanged, freq is ``n_fft // 2 + 1`` where ``n_fft`` is the number of
Fourier bins, and time is the number of window hops (n_frames).
"""
assert waveform.dim() == 2
......@@ -221,7 +223,7 @@ def amplitude_to_DB(x, multiplier, amin, db_multiplier, top_db=None):
amin (float): Number to clamp ``x``
db_multiplier (float): Log10(max(reference value and amin))
top_db (Optional[float]): Minimum negative cut-off in decibels. A reasonable number
is 80.
is 80. (Default: ``None``)
Returns:
torch.Tensor: Output tensor in decibel scale
......@@ -249,11 +251,11 @@ def create_fb_matrix(n_freqs, f_min, f_max, n_mels):
n_mels (int): Number of mel filterbanks
Returns:
torch.Tensor: Triangular filter banks (fb matrix) of size (`n_freqs`, `n_mels`)
torch.Tensor: Triangular filter banks (fb matrix) of size (``n_freqs``, ``n_mels``)
meaning number of frequencies to highlight/apply to x the number of filterbanks.
Each column is a filterbank so that assuming there is a matrix A of
size (..., `n_freqs`), the applied result would be
`A * create_fb_matrix(A.size(-1), ...)`.
size (..., ``n_freqs``), the applied result would be
``A * create_fb_matrix(A.size(-1), ...)``.
"""
# freq bins
freqs = torch.linspace(f_min, f_max, n_freqs)
......@@ -278,7 +280,7 @@ def create_fb_matrix(n_freqs, f_min, f_max, n_mels):
@torch.jit.script
def create_dct(n_mfcc, n_mels, norm):
# type: (int, int, Optional[str]) -> Tensor
r"""Creates a DCT transformation matrix with shape (`n_mels`, `n_mfcc`),
r"""Creates a DCT transformation matrix with shape (``n_mels``, ``n_mfcc``),
normalized depending on norm.
Args:
......@@ -288,7 +290,7 @@ def create_dct(n_mfcc, n_mels, norm):
Returns:
torch.Tensor: The transformation matrix, to be right-multiplied to
row-wise data of size (`n_mels`, `n_mfcc`).
row-wise data of size (``n_mels``, ``n_mfcc``).
"""
# http://en.wikipedia.org/wiki/Discrete_cosine_transform#DCT-II
n = torch.arange(float(n_mels))
......@@ -317,7 +319,7 @@ def mu_law_encoding(x, quantization_channels):
quantization_channels (int): Number of channels
Returns:
torch.Tensor: Input after mu-law companding
torch.Tensor: Input after mu-law encoding
"""
mu = quantization_channels - 1.
if not x.is_floating_point():
......@@ -343,7 +345,7 @@ def mu_law_decoding(x_mu, quantization_channels):
quantization_channels (int): Number of channels
Returns:
torch.Tensor: Input after decoding
torch.Tensor: Input after mu-law decoding
"""
mu = quantization_channels - 1.
if not x_mu.is_floating_point():
......@@ -382,14 +384,14 @@ def angle(complex_tensor):
def magphase(complex_tensor, power=1.):
r"""Separate a complex-valued spectrogram with shape (*,2) into its magnitude and phase.
r"""Separate a complex-valued spectrogram with shape `(*, 2)` into its magnitude and phase.
Args:
complex_tensor (torch.Tensor): Tensor shape of `(*, complex=2)`
power (float): Power of the norm. (Default: `1.0`)
Returns:
Tuple[torch.Tensor, torch.Tensor]: The magnitude and phase of the complex_tensor
Tuple[torch.Tensor, torch.Tensor]: The magnitude and phase of the complex tensor
"""
mag = complex_norm(complex_tensor, power)
phase = angle(complex_tensor)
......@@ -398,17 +400,19 @@ def magphase(complex_tensor, power=1.):
def phase_vocoder(complex_specgrams, rate, phase_advance):
r"""Given a STFT tensor, speed up in time without modifying pitch by a
factor of `rate`.
factor of ``rate``.
Args:
complex_specgrams (torch.Tensor): Size of (*, c, f, t, complex=2)
complex_specgrams (torch.Tensor): Dimension of `(*, channel, freq, time, complex=2)`
rate (float): Speed-up factor
phase_advance (torch.Tensor): Expected phase advance in each bin. Size of (f, 1)
phase_advance (torch.Tensor): Expected phase advance in each bin. Dimension
of (freq, 1)
Returns:
complex_specgrams_stretch (torch.Tensor): Size of (*, c, f, ceil(t/rate), complex=2)
complex_specgrams_stretch (torch.Tensor): Dimension of `(*, channel,
freq, ceil(time/rate), complex=2)`
Example:
Example
>>> num_freqs, hop_length = 1025, 512
>>> # (batch, channel, num_freqs, time, complex=2)
>>> complex_specgrams = torch.randn(16, 1, num_freqs, 300, 2)
......
torchaudio/kaldi_io.py
View file @ 95235f31
......@@ -21,17 +21,18 @@ __all__ = [
def _convert_method_output_to_tensor(file_or_fd, fn, convert_contiguous=False):
r""" Takes a method invokes it. The output is converted to a tensor.
r"""Takes a method invokes it. The output is converted to a tensor.
Arguments:
file_or_fd (string/File Descriptor): file name or file descriptor.
fn (Function): function that has the signature (file name/descriptor) -> generator(string, ndarray)
and converts it to (file name/descriptor) -> generator(string, Tensor).
convert_contiguous (bool): determines whether the array should be converted into a
contiguous layout.
Args:
file_or_fd (str/FileDescriptor): File name or file descriptor
fn (Callable[[...], Generator[str, numpy.ndarray]]): Function that has the signature (
file name/descriptor) -> Generator(str, numpy.ndarray) and converts it to (
file name/descriptor) -> Generator(str, torch.Tensor).
convert_contiguous (bool): Determines whether the array should be converted into a
contiguous layout. (Default: ``None``)
Returns:
generator[key (string), vec/mat (Tensor)]
Generator[str, torch.Tensor]: The string is the key and the tensor is vec/mat
"""
if not IMPORT_KALDI_IO:
raise ImportError('Could not import kaldi_io. Did you install it?')
......@@ -45,14 +46,13 @@ def _convert_method_output_to_tensor(file_or_fd, fn, convert_contiguous=False):
def read_vec_int_ark(file_or_fd):
r"""Create generator of (key,vector<int>) tuples, which reads from the ark file/stream.
Arguments:
file_or_fd (string/File Descriptor): ark, gzipped ark, pipe or opened file descriptor.
Args:
file_or_fd (str/FileDescriptor): ark, gzipped ark, pipe or opened file descriptor
Returns:
generator[key (string), vec (Tensor)]
Example::
Generator[str, torch.Tensor]: The string is the key and the tensor is the vector read from file
Example
>>> # read ark to a 'dictionary'
>>> d = { u:d for u,d in torchaudio.kaldi_io.read_vec_int_ark(file) }
"""
......@@ -63,16 +63,15 @@ def read_vec_int_ark(file_or_fd):
def read_vec_flt_scp(file_or_fd):
r"""Create generator of (key,vector<float32/float64>) tuples, read according to kaldi scp.
r"""Create generator of (key,vector<float32/float64>) tuples, read according to Kaldi scp.
Arguments:
file_or_fd (string/File Descriptor): scp, gzipped scp, pipe or opened file descriptor.
Args:
file_or_fd (str/FileDescriptor): scp, gzipped scp, pipe or opened file descriptor
Returns:
generator[key (string), vec (Tensor)]
Example::
Generator[str, torch.Tensor]: The string is the key and the tensor is the vector read from file
Example
>>> # read scp to a 'dictionary'
>>> # d = { u:d for u,d in torchaudio.kaldi_io.read_vec_flt_scp(file) }
"""
......@@ -82,14 +81,13 @@ def read_vec_flt_scp(file_or_fd):
def read_vec_flt_ark(file_or_fd):
r"""Create generator of (key,vector<float32/float64>) tuples, which reads from the ark file/stream.
Arguments:
file_or_fd (string/File Descriptor): ark, gzipped ark, pipe or opened file descriptor.
Args:
file_or_fd (str/FileDescriptor): ark, gzipped ark, pipe or opened file descriptor
Returns:
generator[key (string), vec (Tensor)]
Example::
Generator[str, torch.Tensor]: The string is the key and the tensor is the vector read from file
Example
>>> # read ark to a 'dictionary'
>>> d = { u:d for u,d in torchaudio.kaldi_io.read_vec_flt_ark(file) }
"""
......@@ -97,16 +95,15 @@ def read_vec_flt_ark(file_or_fd):
def read_mat_scp(file_or_fd):
r"""Create generator of (key,matrix<float32/float64>) tuples, read according to kaldi scp.
r"""Create generator of (key,matrix<float32/float64>) tuples, read according to Kaldi scp.
Arguments:
file_or_fd (string/File Descriptor): scp, gzipped scp, pipe or opened file descriptor.
Args:
file_or_fd (str/FileDescriptor): scp, gzipped scp, pipe or opened file descriptor
Returns:
generator[key (string), mat (Tensor)]
Example::
Generator[str, torch.Tensor]: The string is the key and the tensor is the matrix read from file
Example
>>> # read scp to a 'dictionary'
>>> d = { u:d for u,d in torchaudio.kaldi_io.read_mat_scp(file) }
"""
......@@ -116,14 +113,13 @@ def read_mat_scp(file_or_fd):
def read_mat_ark(file_or_fd):
r"""Create generator of (key,matrix<float32/float64>) tuples, which reads from the ark file/stream.
Arguments:
file_or_fd (string/File Descriptor): ark, gzipped ark, pipe or opened file descriptor.
Args:
file_or_fd (str/FileDescriptor): ark, gzipped ark, pipe or opened file descriptor
Returns:
generator[key (string), mat (Tensor)]
Example::
Generator[str, torch.Tensor]: The string is the key and the tensor is the matrix read from file
Example
>>> # read ark to a 'dictionary'
>>> d = { u:d for u,d in torchaudio.kaldi_io.read_mat_ark(file) }
"""
......
torchaudio/legacy.py
View file @ 95235f31
......@@ -8,51 +8,50 @@ import torchaudio
def load(filepath, out=None, normalization=None, num_frames=0, offset=0):
"""Loads an audio file from disk into a Tensor. The default options have
r"""Loads an audio file from disk into a Tensor. The default options have
changed as of torchaudio 0.2 and this function maintains option defaults
from version 0.1.
Args:
filepath (string): path to audio file
out (Tensor, optional): an output Tensor to use instead of creating one
filepath (str): Path to audio file
out (torch.Tensor, optional): An output Tensor to use instead of creating one. (Default: ``None``)
normalization (bool or number, optional): If boolean `True`, then output is divided by `1 << 31`
(assumes 16-bit depth audio, and normalizes to `[0, 1]`.
If `number`, then output is divided by that number
num_frames (int, optional): number of frames to load. -1 to load everything after the offset.
offset (int, optional): number of frames from the start of the file to begin data loading.
Returns: tuple(Tensor, int)
- Tensor: output Tensor of size `[L x C]` where L is the number of audio frames, C is the number of channels
- int: the sample-rate of the audio (as listed in the metadata of the file)
Example::
(assumes 16-bit depth audio, and normalizes to `[0, 1]`. If `number`, then output is divided by that
number. (Default: ``None``)
num_frames (int, optional): Number of frames to load. -1 to load everything after the
offset. (Default: ``0``)
offset (int, optional): Number of frames from the start of the file to begin data
loading. (Default: ``0``)
Returns:
Tuple[torch.Tensor, int]: The output tensor is of size `[L x C]` where L is the number of audio frames,
C is the number of channels. The integer is sample-rate of the audio (as listed in the metadata of
the file)
Example
>>> data, sample_rate = torchaudio.legacy.load('foo.mp3')
>>> print(data.size())
torch.Size([278756, 2])
>>> print(sample_rate)
44100
"""
return torchaudio.load(filepath, out, normalization, False, num_frames, offset)
def save(filepath, src, sample_rate, precision=32):
"""Saves a Tensor with audio signal to disk as a standard format like mp3, wav, etc.
r"""Saves a Tensor with audio signal to disk as a standard format like mp3, wav, etc.
The default options have changed as of torchaudio 0.2 and this function maintains
option defaults from version 0.1.
Args:
filepath (string): path to audio file
src (Tensor): an input 2D Tensor of shape `[L x C]` where L is
the number of audio frames, C is the number of channels
sample_rate (int): the sample-rate of the audio to be saved
precision (int, optional): the bit-precision of the audio to be saved
Example::
filepath (str): Path to audio file
src (torch.Tensor): An input 2D Tensor of shape `[L x C]` where L is
the number of audio frames, C is the number of channels
sample_rate (int): The sample-rate of the audio to be saved
precision (int, optional): The bit-precision of the audio to be saved. (Default: ``32``)
Example
>>> data, sample_rate = torchaudio.legacy.load('foo.mp3')
>>> torchaudio.legacy.save('foo.wav', data, sample_rate)
"""
torchaudio.save(filepath, src, sample_rate, precision, False)
torchaudio/sox_effects.py
View file @ 95235f31
......@@ -10,61 +10,59 @@ def effect_names():
Returns: list[str]
Example::
Example
>>> EFFECT_NAMES = torchaudio.sox_effects.effect_names()
"""
return _torch_sox.get_effect_names()
def SoxEffect():
"""Create an object for passing sox effect information between python and c++
r"""Create an object for passing sox effect information between python and c++
Returns: SoxEffect(object)
- ename (str), name of effect
- eopts (list[str]), list of effect options
Returns:
SoxEffect: An object with the following attributes: ename (str) which is the
name of effect, and eopts (List[str]) which is a list of effect options.
"""
return _torch_sox.SoxEffect()
class SoxEffectsChain(object):
"""SoX effects chain class.
r"""SoX effects chain class.
Args:
normalization (bool, number, or callable, optional): If boolean `True`, then output is divided by `1 << 31`
(assumes signed 32-bit audio), and normalizes to `[0, 1]`.
If `number`, then output is divided by that number
If `callable`, then the output is passed as a parameter
to the given function, then the output is divided by
the result.
channels_first (bool, optional): Set channels first or length first in result. Default: ``True``
(assumes signed 32-bit audio), and normalizes to `[0, 1]`. If `number`, then output is divided by that
number. If `callable`, then the output is passed as a parameter to the given function, then the
output is divided by the result. (Default: ``True``)
channels_first (bool, optional): Set channels first or length first in result. (Default: ``True``)
out_siginfo (sox_signalinfo_t, optional): a sox_signalinfo_t type, which could be helpful if the
audio type cannot be automatically determined
audio type cannot be automatically determined. (Default: ``None``)
out_encinfo (sox_encodinginfo_t, optional): a sox_encodinginfo_t type, which could be set if the
audio type cannot be automatically determined
filetype (str, optional): a filetype or extension to be set if sox cannot determine it automatically
Returns: tuple(Tensor, int)
- Tensor: output Tensor of size `[C x L]` or `[L x C]` where L is the number of audio frames and
C is the number of channels
- int: the sample rate of the audio (as listed in the metadata of the file)
Example::
class MyDataset(Dataset):
def __init__(self, audiodir_path):
self.data = [os.path.join(audiodir_path, fn) for fn in os.listdir(audiodir_path)]
self.E = torchaudio.sox_effects.SoxEffectsChain()
self.E.append_effect_to_chain("rate", [16000]) # resample to 16000hz
self.E.append_effect_to_chain("channels", ["1"]) # mono signal
def __getitem__(self, index):
fn = self.data[index]
self.E.set_input_file(fn)
x, sr = self.E.sox_build_flow_effects()
return x, sr
def __len__(self):
return len(self.data)
audio type cannot be automatically determined. (Default: ``None``)
filetype (str, optional): a filetype or extension to be set if sox cannot determine it
automatically. . (Default: ``'raw'``)
Returns:
Tuple[torch.Tensor, int]: An output Tensor of size `[C x L]` or `[L x C]` where L is the number
of audio frames and C is the number of channels. An integer which is the sample rate of the
audio (as listed in the metadata of the file)
Example
>>> class MyDataset(Dataset):
>>> def __init__(self, audiodir_path):
>>> self.data = [os.path.join(audiodir_path, fn) for fn in os.listdir(audiodir_path)]
>>> self.E = torchaudio.sox_effects.SoxEffectsChain()
>>> self.E.append_effect_to_chain("rate", [16000]) # resample to 16000hz
>>> self.E.append_effect_to_chain("channels", ["1"]) # mono signal
>>> def __getitem__(self, index):
>>> fn = self.data[index]
>>> self.E.set_input_file(fn)
>>> x, sr = self.E.sox_build_flow_effects()
>>> return x, sr
>>>
>>> def __len__(self):
>>> return len(self.data)
>>>
>>> torchaudio.initialize_sox()
>>> ds = MyDataset(path_to_audio_files)
>>> for sig, sr in ds:
......@@ -87,7 +85,11 @@ class SoxEffectsChain(object):
self.channels_first = channels_first
def append_effect_to_chain(self, ename, eargs=None):
"""Append effect to a sox effects chain.
r"""Append effect to a sox effects chain.
Args:
ename (str): which is the name of effect
eargs (List[str]): which is a list of effect options. (Default: ``None``)
"""
e = SoxEffect()
# check if we have a valid effect
......@@ -106,7 +108,15 @@ class SoxEffectsChain(object):
self.chain.append(e)
def sox_build_flow_effects(self, out=None):
"""Build effects chain and flow effects from input file to output tensor
r"""Build effects chain and flow effects from input file to output tensor
Args:
out (torch.Tensor): Where the output will be written to. (Default: ``None``)
Returns:
Tuple[torch.Tensor, int]: An output Tensor of size `[C x L]` or `[L x C]` where L is the number
of audio frames and C is the number of channels. An integer which is the sample rate of the
audio (as listed in the metadata of the file)
"""
# initialize output tensor
if out is not None:
......@@ -134,12 +144,15 @@ class SoxEffectsChain(object):
return out, sr
def clear_chain(self):
"""Clear effects chain in python
r"""Clear effects chain in python
"""
self.chain = []
def set_input_file(self, input_file):
"""Set input file for input of chain
r"""Set input file for input of chain
Args:
input_file (str): The path to the input file.
"""
self.input_file = input_file
......
torchaudio/transforms.py
View file @ 95235f31
......@@ -23,17 +23,17 @@ class Spectrogram(torch.jit.ScriptModule):
r"""Create a spectrogram from a audio signal
Args:
n_fft (int, optional): Size of fft, creates `n_fft // 2 + 1` bins
win_length (int): Window size. (Default: `n_fft`)
n_fft (int, optional): Size of FFT, creates ``n_fft // 2 + 1`` bins
win_length (int): Window size. (Default: ``n_fft``)
hop_length (int, optional): Length of hop between STFT windows. (
Default: `win_length // 2`)
pad (int): Two sided padding of signal. (Default: 0)
Default: ``win_length // 2``)
pad (int): Two sided padding of signal. (Default: ``0``)
window_fn (Callable[[...], torch.Tensor]): A function to create a window tensor
that is applied/multiplied to each frame/window. (Default: `torch.hann_window`)
that is applied/multiplied to each frame/window. (Default: ``torch.hann_window``)
power (int): Exponent for the magnitude spectrogram,
(must be > 0) e.g., 1 for energy, 2 for power, etc.
normalized (bool): Whether to normalize by magnitude after stft. (Default: `False`)
wkwargs (Dict[..., ...]): Arguments for window function. (Default: `None`)
(must be > 0) e.g., 1 for energy, 2 for power, etc. (Default: ``2``)
normalized (bool): Whether to normalize by magnitude after stft. (Default: ``False``)
wkwargs (Dict[..., ...]): Arguments for window function. (Default: ``None``)
"""
__constants__ = ['n_fft', 'win_length', 'hop_length', 'pad', 'power', 'normalized']
......@@ -42,7 +42,7 @@ class Spectrogram(torch.jit.ScriptModule):
power=2, normalized=False, wkwargs=None):
super(Spectrogram, self).__init__()
self.n_fft = n_fft
# number of fft bins. the returned STFT result will have n_fft // 2 + 1
# number of FFT bins. the returned STFT result will have n_fft // 2 + 1
# number of frequecies due to onesided=True in torch.stft
self.win_length = win_length if win_length is not None else n_fft
self.hop_length = hop_length if hop_length is not None else self.win_length // 2
......@@ -56,12 +56,12 @@ class Spectrogram(torch.jit.ScriptModule):
def forward(self, waveform):
r"""
Args:
waveform (torch.Tensor): Tensor of audio of size (c, n)
waveform (torch.Tensor): Tensor of audio of dimension (channel, time)
Returns:
torch.Tensor: Channels x frequency x time (c, f, t), where channels
is unchanged, frequency is `n_fft // 2 + 1` where `n_fft` is the number of
fourier bins, and time is the number of window hops (n_frames).
torch.Tensor: Dimension (channel, freq, time), where channel
is unchanged, freq is ``n_fft // 2 + 1`` where ``n_fft`` is the number of
Fourier bins, and time is the number of window hops (n_frames).
"""
return F.spectrogram(waveform, self.pad, self.window, self.n_fft, self.hop_length,
self.win_length, self.power, self.normalized)
......@@ -76,9 +76,9 @@ class AmplitudeToDB(torch.jit.ScriptModule):
Args:
stype (str): scale of input tensor ('power' or 'magnitude'). The
power being the elementwise square of the magnitude. (Default: 'power')
power being the elementwise square of the magnitude. (Default: ``'power'``)
top_db (float, optional): minimum negative cut-off in decibels. A reasonable number
is 80.
is 80. (Default: ``None``)
"""
__constants__ = ['multiplier', 'amin', 'ref_value', 'db_multiplier']
......@@ -109,17 +109,17 @@ class AmplitudeToDB(torch.jit.ScriptModule):
class MelScale(torch.jit.ScriptModule):
r"""This turns a normal STFT into a mel frequency STFT, using a conversion
matrix. This uses triangular filter banks.
matrix. This uses triangular filter banks.
User can control which device the filter bank (`fb`) is (e.g. fb.to(spec_f.device)).
User can control which device the filter bank (`fb`) is (e.g. fb.to(spec_f.device)).
Args:
n_mels (int): Number of mel filterbanks. (Default: 128)
sample_rate (int): Sample rate of audio signal. (Default: 16000)
f_min (float): Minimum frequency. (Default: 0.)
f_max (float, optional): Maximum frequency. (Default: `sample_rate // 2`)
n_mels (int): Number of mel filterbanks. (Default: ``128``)
sample_rate (int): Sample rate of audio signal. (Default: ``16000``)
f_min (float): Minimum frequency. (Default: ``0.``)
f_max (float, optional): Maximum frequency. (Default: ``sample_rate // 2``)
n_stft (int, optional): Number of bins in STFT. Calculated from first input
if `None` is given. See `n_fft` in `Spectrogram`.
if None is given. See ``n_fft`` in :class:`Spectrogram`.
"""
__constants__ = ['n_mels', 'sample_rate', 'f_min', 'f_max']
......@@ -138,10 +138,10 @@ class MelScale(torch.jit.ScriptModule):
def forward(self, specgram):
r"""
Args:
specgram (torch.Tensor): a spectrogram STFT of size (c, f, t)
specgram (torch.Tensor): A spectrogram STFT of dimension (channel, freq, time)
Returns:
torch.Tensor: mel frequency spectrogram of size (c, `n_mels`, t)
torch.Tensor: Mel frequency spectrogram of size (channel, ``n_mels``, time)
"""
if self.fb.numel() == 0:
tmp_fb = F.create_fb_matrix(specgram.size(1), self.f_min, self.f_max, self.n_mels)
......@@ -149,7 +149,8 @@ class MelScale(torch.jit.ScriptModule):
self.fb.resize_(tmp_fb.size())
self.fb.copy_(tmp_fb)
# (c, f, t).transpose(...) dot (f, n_mels) -> (c, t, n_mels).transpose(...)
# (channel, frequency, time).transpose(...) dot (frequency, n_mels)
# -> (channel, time, n_mels).transpose(...)
mel_specgram = torch.matmul(specgram.transpose(1, 2), self.fb).transpose(1, 2)
return mel_specgram
......@@ -158,28 +159,28 @@ class MelSpectrogram(torch.jit.ScriptModule):
r"""Create MelSpectrogram for a raw audio signal. This is a composition of Spectrogram
and MelScale.
Sources:
Sources
* https://gist.github.com/kastnerkyle/179d6e9a88202ab0a2fe
* https://timsainb.github.io/spectrograms-mfccs-and-inversion-in-python.html
* http://haythamfayek.com/2016/04/21/speech-processing-for-machine-learning.html
Args:
sample_rate (int): Sample rate of audio signal. (Default: 16000)
win_length (int): Window size. (Default: `n_fft`)
sample_rate (int): Sample rate of audio signal. (Default: ``16000``)
win_length (int): Window size. (Default: ``n_fft``)
hop_length (int, optional): Length of hop between STFT windows. (
Default: `win_length // 2`)
n_fft (int, optional): Size of fft, creates `n_fft // 2 + 1` bins
f_min (float): Minimum frequency. (Default: 0.)
f_max (float, optional): Maximum frequency. (Default: `None`)
pad (int): Two sided padding of signal. (Default: 0)
n_mels (int): Number of mel filterbanks. (Default: 128)
Default: ``win_length // 2``)
n_fft (int, optional): Size of FFT, creates ``n_fft // 2 + 1`` bins
f_min (float): Minimum frequency. (Default: ``0.``)
f_max (float, optional): Maximum frequency. (Default: ``None``)
pad (int): Two sided padding of signal. (Default: ``0``)
n_mels (int): Number of mel filterbanks. (Default: ``128``)
window_fn (Callable[[...], torch.Tensor]): A function to create a window tensor
that is applied/multiplied to each frame/window. (Default: `torch.hann_window`)
wkwargs (Dict[..., ...]): Arguments for window function. (Default: `None`)
that is applied/multiplied to each frame/window. (Default: ``torch.hann_window``)
wkwargs (Dict[..., ...]): Arguments for window function. (Default: ``None``)
Example:
Example
>>> waveform, sample_rate = torchaudio.load('test.wav', normalization=True)
>>> mel_specgram = transforms.MelSpectrogram(sample_rate)(waveform) # (c, n_mels, t)
>>> mel_specgram = transforms.MelSpectrogram(sample_rate)(waveform) # (channel, n_mels, time)
"""
__constants__ = ['sample_rate', 'n_fft', 'win_length', 'hop_length', 'pad', 'n_mels', 'f_min']
......@@ -204,10 +205,10 @@ class MelSpectrogram(torch.jit.ScriptModule):
def forward(self, waveform):
r"""
Args:
waveform (torch.Tensor): Tensor of audio of size (c, n)
waveform (torch.Tensor): Tensor of audio of dimension (channel, time)
Returns:
torch.Tensor: mel frequency spectrogram of size (c, `n_mels`, t)
torch.Tensor: Mel frequency spectrogram of size (channel, ``n_mels``, time)
"""
specgram = self.spectrogram(waveform)
mel_specgram = self.mel_scale(specgram)
......@@ -226,12 +227,13 @@ class MFCC(torch.jit.ScriptModule):
a full clip.
Args:
sample_rate (int): Sample rate of audio signal. (Default: 16000)
n_mfcc (int): Number of mfc coefficients to retain
dct_type (int): type of DCT (discrete cosine transform) to use
norm (string, optional): norm to use
log_mels (bool): whether to use log-mel spectrograms instead of db-scaled
melkwargs (dict, optional): arguments for MelSpectrogram
sample_rate (int): Sample rate of audio signal. (Default: ``16000``)
n_mfcc (int): Number of mfc coefficients to retain. (Default: ``40``)
dct_type (int): type of DCT (discrete cosine transform) to use. (Default: ``2``)
norm (str, optional): norm to use. (Default: ``'ortho'``)
log_mels (bool): whether to use log-mel spectrograms instead of db-scaled. (Default:
``False``)
melkwargs (dict, optional): arguments for MelSpectrogram. (Default: ``None``)
"""
__constants__ = ['sample_rate', 'n_mfcc', 'dct_type', 'top_db', 'log_mels']
......@@ -263,10 +265,10 @@ class MFCC(torch.jit.ScriptModule):
def forward(self, waveform):
r"""
Args:
waveform (torch.Tensor): Tensor of audio of size (c, n)
waveform (torch.Tensor): Tensor of audio of dimension (channel, time)
Returns:
torch.Tensor: specgram_mel_db of size (c, `n_mfcc`, t)
torch.Tensor: specgram_mel_db of size (channel, ``n_mfcc``, time)
"""
mel_specgram = self.MelSpectrogram(waveform)
if self.log_mels:
......@@ -274,7 +276,8 @@ class MFCC(torch.jit.ScriptModule):
mel_specgram = torch.log(mel_specgram + log_offset)
else:
mel_specgram = self.amplitude_to_DB(mel_specgram)
# (c, `n_mels`, t).tranpose(...) dot (`n_mels`, `n_mfcc`) -> (c, t, `n_mfcc`).tranpose(...)
# (channel, n_mels, time).tranpose(...) dot (n_mels, n_mfcc)
# -> (channel, time, n_mfcc).tranpose(...)
mfcc = torch.matmul(mel_specgram.transpose(1, 2), self.dct_mat).transpose(1, 2)
return mfcc
......@@ -287,7 +290,7 @@ class MuLawEncoding(torch.jit.ScriptModule):
returns a signal encoded with values from 0 to quantization_channels - 1
Args:
quantization_channels (int): Number of channels (Default: 256)
quantization_channels (int): Number of channels (Default: ``256``)
"""
__constants__ = ['quantization_channels']
......@@ -315,7 +318,7 @@ class MuLawDecoding(torch.jit.ScriptModule):
and returns a signal scaled between -1 and 1.
Args:
quantization_channels (int): Number of channels (Default: 256)
quantization_channels (int): Number of channels (Default: ``256``)
"""
__constants__ = ['quantization_channels']
......@@ -340,11 +343,11 @@ class Resample(torch.nn.Module):
be given.
Args:
orig_freq (float): The original frequency of the signal
new_freq (float): The desired frequency
resampling_method (str): The resampling method (Default: 'sinc_interpolation')
orig_freq (float): The original frequency of the signal. (Default: ``16000``)
new_freq (float): The desired frequency. (Default: ``16000``)
resampling_method (str): The resampling method (Default: ``'sinc_interpolation'``)
"""
def __init__(self, orig_freq, new_freq, resampling_method='sinc_interpolation'):
def __init__(self, orig_freq=16000, new_freq=16000, resampling_method='sinc_interpolation'):
super(Resample, self).__init__()
self.orig_freq = orig_freq
self.new_freq = new_freq
......@@ -353,10 +356,10 @@ class Resample(torch.nn.Module):
def forward(self, waveform):
r"""
Args:
waveform (torch.Tensor): The input signal of size (c, n)
waveform (torch.Tensor): The input signal of dimension (channel, time)
Returns:
torch.Tensor: Output signal of size (c, m)
torch.Tensor: Output signal of dimension (channel, time)
"""
if self.resampling_method == 'sinc_interpolation':
return kaldi.resample_waveform(waveform, self.orig_freq, self.new_freq)
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment