Skip to content

Snr db #14

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wesbz wants to merge 2 commits into
base: master
Choose a base branch
from
Open

Snr db #14

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
97f7c1a
Expressing SNR in dB.
wesbz Jan 9, 2022
6bae117
Correcting the values of SNR
wesbz Jan 9, 2022
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,7 +27,7 @@ num_samples = sr * 5
transforms = [
RandomResizedCrop(n_samples=num_samples),
RandomApply([PolarityInversion()], p=0.8),
RandomApply([Noise(min_snr=0.001, max_snr=0.005)], p=0.3),
RandomApply([Noise(min_snr=-60, max_snr=-46)], p=0.3),
RandomApply([Gain()], p=0.2),
HighLowPass(sample_rate=sr), # this augmentation will always be applied in this aumgentation chain!
RandomApply([Delay(sample_rate=sr)], p=0.5),
Expand All @@ -43,7 +43,7 @@ We can also define a stochastic augmentation on multiple transformations. The fo
```python
transforms = [
RandomResizedCrop(n_samples=num_samples),
RandomApply([PolarityInversion(), Noise(min_snr=0.001, max_snr=0.005)], p=0.8),
RandomApply([PolarityInversion(), Noise(min_snr=-60, max_snr=-46)], p=0.8),
RandomApply([Gain()], p=0.2),
RandomApply([Delay(sample_rate=sr), Reverb(sample_rate=sr)], p=0.5)
]
Expand Down
2 changes: 1 addition & 1 deletion tests/test_augmentations.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -179,7 +179,7 @@ def test_gain(num_channels):
def test_noise(num_channels):
audio = generate_waveform(sample_rate, num_samples, num_channels)
transform = Compose(
[Noise(min_snr=0.5, max_snr=1)],
[Noise(min_snr=-6, max_snr=0)],
)

t_audio = transform(audio)
Expand Down
2 changes: 1 addition & 1 deletion tests/test_readme_example.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,7 +29,7 @@ def test_readme_example():
transforms = [
RandomResizedCrop(n_samples=num_samples),
RandomApply([PolarityInversion()], p=0.8),
RandomApply([Noise(min_snr=0.3, max_snr=0.5)], p=0.3),
RandomApply([Noise(min_snr=-10, max_snr=-6)], p=0.3),
RandomApply([Gain()], p=0.2),
RandomApply([HighLowPass(sample_rate=sr)], p=0.8),
RandomApply([Delay(sample_rate=sr)], p=0.5),
Expand Down
9 changes: 5 additions & 4 deletions torchaudio_augmentations/augmentations/noise.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,18 +4,19 @@


class Noise(torch.nn.Module):
def __init__(self, min_snr=0.0001, max_snr=0.01):
def __init__(self, min_snr=-80, max_snr=-40):
"""
:param min_snr: Minimum signal-to-noise ratio
:param max_snr: Maximum signal-to-noise ratio
:param min_snr: Minimum signal-to-noise ratio in dB.
:param max_snr: Maximum signal-to-noise ratio in dB.
"""
super().__init__()
self.min_snr = min_snr
self.max_snr = max_snr

def forward(self, audio):
std = torch.std(audio)
noise_std = random.uniform(self.min_snr * std, self.max_snr * std)
snr = random.uniform(self.min_snr, self.max_snr)
noise_std = 10**(snr/20) * std

noise = np.random.normal(0.0, noise_std, size=audio.shape).astype(np.float32)

Expand Down