• Docs >
  • Audio Data Augmentation >
  • Nightly (unstable)

Audio Data Augmentation

Author: Moto Hira

torchaudio provides a variety of ways to augment audio data.

In this tutorial, we look into a way to apply effects, filters, RIR (room impulse response) and codecs.

At the end, we synthesize noisy speech over phone from clean speech.

import torch
import torchaudio
import torchaudio.functional as F


import matplotlib.pyplot as plt


First, we import the modules and download the audio assets we use in this tutorial.

from IPython.display import Audio

from torchaudio.utils import download_asset

SAMPLE_WAV = download_asset("tutorial-assets/steam-train-whistle-daniel_simon.wav")
SAMPLE_RIR = download_asset("tutorial-assets/Lab41-SRI-VOiCES-rm1-impulse-mc01-stu-clo-8000hz.wav")
SAMPLE_SPEECH = download_asset("tutorial-assets/Lab41-SRI-VOiCES-src-sp0307-ch127535-sg0042-8000hz.wav")
SAMPLE_NOISE = download_asset("tutorial-assets/Lab41-SRI-VOiCES-rm1-babb-mc01-stu-clo-8000hz.wav")
  0%|          | 0.00/427k [00:00<?, ?B/s]
100%|##########| 427k/427k [00:00<00:00, 180MB/s]

  0%|          | 0.00/31.3k [00:00<?, ?B/s]
100%|##########| 31.3k/31.3k [00:00<00:00, 57.8MB/s]

  0%|          | 0.00/78.2k [00:00<?, ?B/s]
100%|##########| 78.2k/78.2k [00:00<00:00, 129MB/s]

Applying effects and filtering

torchaudio.io.AudioEffector allows for directly applying filters and codecs to Tensor objects, in a similar way as ffmpeg command

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_rate = torchaudio.load(SAMPLE_WAV, channels_first=False)

# Define effects
effect = ",".join(
        "lowpass=frequency=300:poles=1",  # apply single-pole lowpass filter
        "atempo=0.8",  # reduce the speed
        # Applying echo gives some dramatic feeling

# Apply 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_rate)
print(waveform2.shape, sample_rate)
torch.Size([109368, 2]) 44100
torch.Size([144642, 2]) 44100

Note that the number of frames and number of channels are different from those of the original after the effects are applied. Let’s listen to the audio.

def plot_waveform(waveform, sample_rate, title="Waveform", xlim=None):
    waveform = waveform.numpy()

    num_channels, num_frames = waveform.shape
    time_axis = torch.arange(0, num_frames) / sample_rate

    figure, axes = plt.subplots(num_channels, 1)
    if num_channels == 1:
        axes = [axes]
    for c in range(num_channels):
        axes[c].plot(time_axis, waveform[c], linewidth=1)
        if num_channels > 1:
            axes[c].set_ylabel(f"Channel {c+1}")
        if xlim:
def plot_specgram(waveform, sample_rate, title="Spectrogram", xlim=None):
    waveform = waveform.numpy()

    num_channels, _ = waveform.shape

    figure, axes = plt.subplots(num_channels, 1)
    if num_channels == 1:
        axes = [axes]
    for c in range(num_channels):
        axes[c].specgram(waveform[c], Fs=sample_rate)
        if num_channels > 1:
            axes[c].set_ylabel(f"Channel {c+1}")
        if xlim:


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)
  • Original
  • Original

Effects applied

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)
  • Effects Applied
  • Effects Applied