search

iver56 / audiomentations

A Python library for audio data augmentation. Inspired by albumentations. Useful for machine learning.
https://iver56.github.io/audiomentations/
MIT License
1.83k stars 187 forks source link

Change default noise_rms="relative_to_whole_input" in AddShortNoises #224

Closed iver56 closed 1 year ago

iver56 commented 2 years ago

The default value for noise_rms is currently "relative", but "relative_to_whole_input" might actually be a better default, given that the loudness of the given signal can vary from very silent to very loud, and "relative" doesn't allow placing any loud short noises in part that is silent in the input

What do you think @Cangonin?

Cangonin commented 2 years ago

I agree!

atamazian commented 1 year ago

This issue is solved, so I suggest to close it.

iver56 commented 1 year ago

I can close this issue after we make the final switch. For now it just says this:

            warnings.warn(
                "The default value of noise_rms in AddShortNoises will change from"
                " 'relative' to 'relative_to_whole_input' in a future version of"
                " audiomentations. If you set the value explicitly, you will not get"
                " this warning.",
                FutureWarning
            )

It would be good to make the switch in the next version 👍