Caroline Chen
Caroline Chen
Resolves #6 Summary of Changes - used the latest version of torchaudio (`0.9.0`), which contains various [resampling improvements](https://github.com/pytorch/audio/issues/1487) - replaced torchaudio `resample_waveform` with the new `functional.resample`, since `compliance.kaldi.resample` is being...
Hi @jonashaag, thanks for compiling and updating this notebook on resampling libraries! However, I noticed there was a small bug in the new torchaudio examples, and also wanted to update...
additionally test autograd for the various reduction parameters for rnnt loss. epsilon needed to be increased slightly for one of the tests.
improve 2 cases of _get_env_from, which is expected to return an `Env` type: 1. input env = "base_env" -- falls through all the if statements and returns the string "base_env"...
trying to use existing miniconda base for docker clusters for testing conda envs. current running into some dependency/version conflicts (colorama, crypto) leading to erroring when restarting runhouse on the cluster.