Fix calculation of SNR value in tutorial #2285
Conversation
|
Hi @hagenw! Thank you for your pull request and welcome to our community. Action RequiredIn order to merge any pull request (code, docs, etc.), we require contributors to sign our Contributor License Agreement, and we don't seem to have one on file for you. ProcessIn order for us to review and merge your suggested changes, please sign at https://code.facebook.com/cla. If you are contributing on behalf of someone else (eg your employer), the individual CLA may not be sufficient and your employer may need to sign the corporate CLA. Once the CLA is signed, our tooling will perform checks and validations. Afterwards, the pull request will be tagged with If you have received this in error or have any questions, please contact us at cla@fb.com. Thanks! |
hagenw
force-pushed
the
fix-snr-calculation
branch
from
c47fef5 to
34e443e
Compare
hagenw
changed the title
|
@mthrok has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
Thank you for signing our Contributor License Agreement. We can now accept your code for this (and any) Meta Open Source project. Thanks! |
|
Hey @mthrok. |
Summary: The calculation of the SNR in tha data augmentation examples seems to be wrong to me:  If we start from the definition of the signal-to-noise ratio using the root mean square value we get: ``` SNR = 20 log10 ( rms(scale * speech) / rms(noise) ) ``` this can be transformed to ``` scale = 10^(SNR/20) rms(noise) / rms(speech) ``` In the example not `rms` is used but `lambda x: x.norm(p=2)`, but as we have the same length of the speech and noise signal, we have ``` rms(noise) / rms(speech) = noise.norm(p=2) / speech.norm(p=2) ``` this would lead us to: ``` 10^(SNR/20) = e^(SNR / 10) ``` which is not true. Hence I changed `e^(SNR / 10)` to `10^(SNR/20)`. For the proposed SNR values of 20 dB, 10 dB, 3 dB the value of the scale would change from 7.39, 2.72, 1.35 to 10.0, 3.16, 1.41. Pull Request resolved: pytorch#2285 Reviewed By: nateanl Differential Revision: D35047737 Pulled By: mthrok fbshipit-source-id: ac24c8fd48ef06b4b611e35163084644330a3ef3
In pytorch#2285, the SNR calculation was fixed, but there was still one that was not fixed. This commit fixes it. Also following the feedback pytorch/tutorials#1930 (comment), update the variable name.
Summary: In #2285, the SNR calculation was fixed, but there was still one that was not fixed. This commit fixes it. Also following the feedback pytorch/tutorials#1930 (comment), update the variable name. Pull Request resolved: #2595 Reviewed By: carolineechen Differential Revision: D38314672 Pulled By: mthrok fbshipit-source-id: b2015e2709729190d97264aa191651b3af4ba856
Summary: In #2285, the SNR calculation was fixed, but there was still one that was not fixed. This commit fixes it. Also following the feedback pytorch/tutorials#1930 (comment), update the variable name. Pull Request resolved: #2595 Reviewed By: carolineechen Differential Revision: D38314672 Pulled By: mthrok fbshipit-source-id: b2015e2709729190d97264aa191651b3af4ba856
The calculation of the SNR in tha data augmentation examples seems to be wrong to me:
If we start from the definition of the signal-to-noise ratio using the root mean square value we get:
this can be transformed to
In the example not
rmsis used butlambda x: x.norm(p=2), but as we have the same length of the speech and noise signal, we havethis would lead us to:
which is not true.
Hence I changed
e^(SNR / 10)to10^(SNR/20).For the proposed SNR values of 20 dB, 10 dB, 3 dB the value of
scalewould change from 7.39, 2.72, 1.35 to 10.0, 3.16, 1.41.