Update data augmentation tutorial (#3375)

Summary:
Replace sox_effects with `torchaudio.io.AudioEffector`

1. To show case the new and better feature
2. To prepare for the upcoming removal of file-like support object

Pull Request resolved: https://github.com/pytorch/audio/pull/3375

Reviewed By: nateanl

Differential Revision: D46379016

Pulled By: mthrok

fbshipit-source-id: 70f24b62494204949f327f6ac6c49f315c9ee315

examples/tutorials/audio_data_augmentation_tutorial.py

@@ -27,8 +27,6 @@ print(torchaudio.__version__)
 # First, we import the modules and download the audio assets we use in this tutorial.
 #
 
-import math
-
 from IPython.display import Audio
 import matplotlib.pyplot as plt
 
@@ -44,56 +42,38 @@ SAMPLE_NOISE = download_asset("tutorial-assets/Lab41-SRI-VOiCES-rm1-babb-mc01-st
 # Applying effects and filtering
 # ------------------------------
 #
-# :py:func:`torchaudio.sox_effects` allows for directly applying filters similar to
-# those available in ``sox`` to Tensor objects and file object audio sources.
-#
-# There are two functions for this:
-#
-# -  :py:func:`torchaudio.sox_effects.apply_effects_tensor` for applying effects
-#    to Tensor.
-# -  :py:func:`torchaudio.sox_effects.apply_effects_file` for applying effects to
-#    other audio sources.
-#
-# Both functions accept effect definitions in the form
-# ``List[List[str]]``.
-# This is mostly consistent with how ``sox`` command works, but one caveat is
-# that ``sox`` adds some effects automatically, whereas ``torchaudio``’s
-# implementation does not.
-#
-# For the list of available effects, please refer to `the sox
-# documentation <http://sox.sourceforge.net/sox.html>`__.
+# :py:class:`torchaudio.io.AudioEffector` allows for directly applying
+# filters and codecs to Tensor objects, in a similar way as ``ffmpeg``
+# command
 #
-# **Tip** If you need to load and resample your audio data on the fly,
-# then you can use :py:func:`torchaudio.sox_effects.apply_effects_file`
-# with effect ``"rate"``.
-#
-# **Note** :py:func:`torchaudio.sox_effects.apply_effects_file` accepts a
-# file-like object or path-like object.
-# Similar to :py:func:`torchaudio.load`, when the audio format cannot be
-# inferred from either the file extension or header, you can provide
-# argument ``format`` to specify the format of the audio source.
-#
-# **Note** This process is not differentiable.
+# `AudioEffector Usages <./effector_tutorial.html>` explains how to use
+# this class, so for the detail, please refer to the tutorial.
 #
 
 # Load the data
-waveform1, sample_rate1 = torchaudio.load(SAMPLE_WAV)
+waveform1, sample_rate = torchaudio.load(SAMPLE_WAV, channels_first=False)
 
 # Define effects
-effects = [
-    ["lowpass", "-1", "300"],  # apply single-pole lowpass filter
-    ["speed", "0.8"],  # reduce the speed
-    # This only changes sample rate, so it is necessary to
-    # add `rate` effect with original sample rate after this.
-    ["rate", f"{sample_rate1}"],
-    ["reverb", "-w"],  # Reverbration gives some dramatic feeling
-]
+effect = ",".join(
+    [
+        "lowpass=frequency=300:poles=1",  # apply single-pole lowpass filter
+        "atempo=0.8",  # reduce the speed
+        "aecho=in_gain=0.8:out_gain=0.9:delays=200:decays=0.3|delays=400:decays=0.3"
+        # Applying echo gives some dramatic feeling
+    ],
+)
+
 
 # Apply effects
-waveform2, sample_rate2 = torchaudio.sox_effects.apply_effects_tensor(waveform1, sample_rate1, effects)
+def apply_effect(waveform, sample_rate, effect):
+    effector = torchaudio.io.AudioEffector(effect=effect)
+    return effector.apply(waveform, sample_rate)
+
+
+waveform2 = apply_effect(waveform1, sample_rate, effect)
 
-print(waveform1.shape, sample_rate1)
-print(waveform2.shape, sample_rate2)
+print(waveform1.shape, sample_rate)
+print(waveform2.shape, sample_rate)
 
 ######################################################################
 # Note that the number of frames and number of channels are different from
@@ -101,6 +81,7 @@ print(waveform2.shape, sample_rate2)
 # audio.
 #
 
+
 def plot_waveform(waveform, sample_rate, title="Waveform", xlim=None):
     waveform = waveform.numpy()
 
@@ -123,6 +104,7 @@ def plot_waveform(waveform, sample_rate, title="Waveform", xlim=None):
 ######################################################################
 #
 
+
 def plot_specgram(waveform, sample_rate, title="Spectrogram", xlim=None):
     waveform = waveform.numpy()
 
@@ -141,26 +123,23 @@ def plot_specgram(waveform, sample_rate, title="Spectrogram", xlim=None):
     plt.show(block=False)
 
 ######################################################################
-# Original:
-# ~~~~~~~~~
+# Original
+# ~~~~~~~~
 #
 
-plot_waveform(waveform1, sample_rate1, title="Original", xlim=(-0.1, 3.2))
-plot_specgram(waveform1, sample_rate1, title="Original", xlim=(0, 3.04))
-Audio(waveform1, rate=sample_rate1)
+plot_waveform(waveform1.T, sample_rate, title="Original", xlim=(-0.1, 3.2))
+plot_specgram(waveform1.T, sample_rate, title="Original", xlim=(0, 3.04))
+Audio(waveform1.T, rate=sample_rate)
 
 ######################################################################
-# Effects applied:
-# ~~~~~~~~~~~~~~~~
+# Effects applied
+# ~~~~~~~~~~~~~~~
 #
 
-plot_waveform(waveform2, sample_rate2, title="Effects Applied", xlim=(-0.1, 3.2))
-plot_specgram(waveform2, sample_rate2, title="Effects Applied", xlim=(0, 3.04))
-Audio(waveform2, rate=sample_rate2)
+plot_waveform(waveform2.T, sample_rate, title="Effects Applied", xlim=(-0.1, 3.2))
+plot_specgram(waveform2.T, sample_rate, title="Effects Applied", xlim=(0, 3.04))
+Audio(waveform2.T, rate=sample_rate)
 
-######################################################################
-# Doesn’t it sound more dramatic?
-#
 
 ######################################################################
 # Simulating room reverberation
@@ -203,8 +182,8 @@ speech, _ = torchaudio.load(SAMPLE_SPEECH)
 augmented = F.fftconvolve(speech, rir)
 
 ######################################################################
-# Original:
-# ~~~~~~~~~
+# Original
+# ~~~~~~~~
 #
 
 plot_waveform(speech, sample_rate, title="Original")
@@ -212,8 +191,8 @@ plot_specgram(speech, sample_rate, title="Original")
 Audio(speech, rate=sample_rate)
 
 ######################################################################
-# RIR applied:
-# ~~~~~~~~~~~~
+# RIR applied
+# ~~~~~~~~~~~
 #
 
 plot_waveform(augmented, sample_rate, title="RIR Applied")
@@ -248,8 +227,8 @@ noisy_speeches = F.add_noise(speech, noise, snr_dbs)
 
 
 ######################################################################
-# Background noise:
-# ~~~~~~~~~~~~~~~~~
+# Background noise
+# ~~~~~~~~~~~~~~~~
 #
 
 plot_waveform(noise, sample_rate, title="Background noise")
@@ -257,8 +236,8 @@ plot_specgram(noise, sample_rate, title="Background noise")
 Audio(noise, rate=sample_rate)
 
 ######################################################################
-# SNR 20 dB:
-# ~~~~~~~~~~
+# SNR 20 dB
+# ~~~~~~~~~
 #
 
 snr_db, noisy_speech = snr_dbs[0], noisy_speeches[0:1]
@@ -267,8 +246,8 @@ plot_specgram(noisy_speech, sample_rate, title=f"SNR: {snr_db} [dB]")
 Audio(noisy_speech, rate=sample_rate)
 
 ######################################################################
-# SNR 10 dB:
-# ~~~~~~~~~~
+# SNR 10 dB
+# ~~~~~~~~~
 #
 
 snr_db, noisy_speech = snr_dbs[1], noisy_speeches[1:2]
@@ -277,8 +256,8 @@ plot_specgram(noisy_speech, sample_rate, title=f"SNR: {snr_db} [dB]")
 Audio(noisy_speech, rate=sample_rate)
 
 ######################################################################
-# SNR 3 dB:
-# ~~~~~~~~~
+# SNR 3 dB
+# ~~~~~~~~
 #
 
 snr_db, noisy_speech = snr_dbs[2], noisy_speeches[2:3]
@@ -291,60 +270,56 @@ Audio(noisy_speech, rate=sample_rate)
 # Applying codec to Tensor object
 # -------------------------------
 #
-# :py:func:`torchaudio.functional.apply_codec` can apply codecs to
+# :py:class:`torchaudio.io.AudioEffector` can also apply codecs to
 # a Tensor object.
 #
-# **Note** This process is not differentiable.
-#
+
+waveform, sample_rate = torchaudio.load(SAMPLE_SPEECH, channels_first=False)
 
 
-waveform, sample_rate = torchaudio.load(SAMPLE_SPEECH)
+def apply_codec(waveform, sample_rate, format, encoder=None):
+    encoder = torchaudio.io.AudioEffector(format=format, encoder=encoder)
+    return encoder.apply(waveform, sample_rate)
 
-configs = [
-    {"format": "wav", "encoding": "ULAW", "bits_per_sample": 8},
-    {"format": "gsm"},
-    {"format": "vorbis", "compression": -1},
-]
-waveforms = []
-for param in configs:
-    augmented = F.apply_codec(waveform, sample_rate, **param)
-    waveforms.append(augmented)
 
 ######################################################################
-# Original:
-# ~~~~~~~~~
+# Original
+# ~~~~~~~~
 #
 
-plot_waveform(waveform, sample_rate, title="Original")
-plot_specgram(waveform, sample_rate, title="Original")
-Audio(waveform, rate=sample_rate)
+plot_waveform(waveform.T, sample_rate, title="Original")
+plot_specgram(waveform.T, sample_rate, title="Original")
+Audio(waveform.T, rate=sample_rate)
 
 ######################################################################
-# 8 bit mu-law:
-# ~~~~~~~~~~~~~
+# 8 bit mu-law
+# ~~~~~~~~~~~~
 #
 
-plot_waveform(waveforms[0], sample_rate, title="8 bit mu-law")
-plot_specgram(waveforms[0], sample_rate, title="8 bit mu-law")
-Audio(waveforms[0], rate=sample_rate)
+mulaw = apply_codec(waveform, sample_rate, "wav", encoder="pcm_mulaw")
+plot_waveform(mulaw.T, sample_rate, title="8 bit mu-law")
+plot_specgram(mulaw.T, sample_rate, title="8 bit mu-law")
+Audio(mulaw.T, rate=sample_rate)
 
 ######################################################################
-# GSM-FR:
-# ~~~~~~~
+# G.722
+# ~~~~~
 #
 
-plot_waveform(waveforms[1], sample_rate, title="GSM-FR")
-plot_specgram(waveforms[1], sample_rate, title="GSM-FR")
-Audio(waveforms[1], rate=sample_rate)
+g722 = apply_codec(waveform, sample_rate, "g722")
+plot_waveform(g722.T, sample_rate, title="G.722")
+plot_specgram(g722.T, sample_rate, title="G.722")
+Audio(g722.T, rate=sample_rate)
 
 ######################################################################
-# Vorbis:
-# ~~~~~~~
+# Vorbis
+# ~~~~~~
 #
 
-plot_waveform(waveforms[2], sample_rate, title="Vorbis")
-plot_specgram(waveforms[2], sample_rate, title="Vorbis")
-Audio(waveforms[2], rate=sample_rate)
+vorbis = apply_codec(waveform, sample_rate, "ogg", encoder="vorbis")
+plot_waveform(vorbis.T, sample_rate, title="Vorbis")
+plot_specgram(vorbis.T, sample_rate, title="Vorbis")
+Audio(vorbis.T, rate=sample_rate)
 
 ######################################################################
 # Simulating a phone recoding
@@ -378,62 +353,52 @@ bg_added = F.add_noise(rir_applied, noise, snr_db)
 plot_specgram(bg_added, sample_rate, title="BG noise added")
 
 # Apply filtering and change sample rate
-filtered, sample_rate2 = torchaudio.sox_effects.apply_effects_tensor(
-    bg_added,
-    sample_rate,
-    effects=[
-        ["lowpass", "4000"],
-        [
-            "compand",
-            "0.02,0.05",
-            "-60,-60,-30,-10,-20,-8,-5,-8,-2,-8",
-            "-8",
-            "-7",
-            "0.05",
-        ],
-        ["rate", "8000"],
-    ],
-)
+effect = ",".join([
+    "lowpass=frequency=4000:poles=1",
+    "compand=attacks=0.02:decays=0.05:points=-60/-60|-30/-10|-20/-8|-5/-8|-2/-8:gain=-8:volume=-7:delay=0.05",
+])
 
-plot_specgram(filtered, sample_rate2, title="Filtered")
+filtered = apply_effect(bg_added.T, sample_rate, effect)
+sample_rate2 = 8000
 
-# Apply telephony codec
-codec_applied = F.apply_codec(filtered, sample_rate2, format="gsm")
+plot_specgram(filtered.T, sample_rate2, title="Filtered")
 
-plot_specgram(codec_applied, sample_rate2, title="GSM Codec Applied")
+# Apply telephony codec
+codec_applied = apply_codec(filtered, sample_rate2, "g722")
+plot_specgram(codec_applied.T, sample_rate2, title="G.722 Codec Applied")
 
 
 ######################################################################
-# Original speech:
-# ~~~~~~~~~~~~~~~~
+# Original speech
+# ~~~~~~~~~~~~~~~
 #
 
 Audio(original_speech, rate=sample_rate)
 
 ######################################################################
-# RIR applied:
-# ~~~~~~~~~~~~
+# RIR applied
+# ~~~~~~~~~~~
 #
 
 Audio(rir_applied, rate=sample_rate)
 
 ######################################################################
-# Background noise added:
-# ~~~~~~~~~~~~~~~~~~~~~~~
+# Background noise added
+# ~~~~~~~~~~~~~~~~~~~~~~
 #
 
 Audio(bg_added, rate=sample_rate)
 
 ######################################################################
-# Filtered:
-# ~~~~~~~~~
+# Filtered
+# ~~~~~~~~
 #
 
-Audio(filtered, rate=sample_rate2)
+Audio(filtered.T, rate=sample_rate2)
 
 ######################################################################
-# Codec applied:
-# ~~~~~~~~~~~~~~
+# Codec applied
+# ~~~~~~~~~~~~~
 #
 
-Audio(codec_applied, rate=sample_rate2)
+Audio(codec_applied.T, rate=sample_rate2)