Harry Coultas Blum
Harry Coultas Blum
It would be nice not to have to disable the extension with several clicks but to do so with a hotkey? - press CTRL+ALT+V then vimium is disabled temporarily -...
Could be a useful feature to simulate a more realistic silence. Perhaps having real audio samples of real noise to put as the "silence" instead of just 0's
For the `tfm_pad_with_silence` method, I don't think this is the correct name. This is more like cropping the audio. I thought this meant adding silence to either end of the...
Currently supported in torchaudio using sox. ```py import torch import torchaudio from pyannote.audio.utils.preview import listen from IPython.display import Audio x = torchaudio.sox_effects.effect_names() effects = [["reverb"]] wave,sr = torchaudio.load("./tests/data/tst00.wav") display(Audio(wave, rate=sr))...
In pyannote.audio we have a a set of functions that run before commiting. This is through the pre-commit git hook. It makes code styling more consistent, checks for missing imports...
Originally [posted](https://github.com/asteroid-team/torch-audiomentations/pull/51#issuecomment-736407087) by @iver56
Create a demonstration of using fastaudio with https://github.com/heartexlabs/label-studio
As pointed out by @jcaw in #85, this would improve the testing as the values in the tensors can't be exactly the same. This will prevent unnecessary CI failures
When this becomes available https://github.com/nteract/papermill/issues/429, implement it
Demonstrate how users of the library could train a model to learn speaker identities.