Katsuya Iida

Preprocessed data are in the wrong path 512/wikipedia_pretrain.

2

`BERT_pretrain.ipynb` instructs to download https://bertonazuremlwestus2.blob.core.windows.net/public/bert_data.tar.gz for the preprocessed data. The tar file contains data in `512/wikipedia_pretrain`, but it should be `512/wiki_pretrain`.

Can you update .NET to .NET Core 3.1 or .NET 6.0? .NET Core 2.2 is end of support. https://dotnet.microsoft.com/en-us/platform/support/policy/dotnet-core The Docker image for .NET Core 2.2 is Debian stretch based...

- [librosa v0.9.0 implements keyword only arguments throughout the package.](https://github.com/librosa/librosa/pull/1431) Calling `librosa.filters.mel()` needs keywords `sr`, `n_fft`, etc. The order of keywords is the same as v0.8.1, so should be compatible...

`asr_en_phone_base-20230314` has 4 layer convolusion. Compare with 2 layer convolusion.

Loading pre-trained models (`torch.hub.load_state_dict_from_url()` or `torch.hub.load()`)

27

@NiklasGustafsson @ericstj @GeorgeS2019 Probably people want to have a place to download pre-trained model files so that they can fine-tune pre-made models by TorchVision and TorchAudio without converting files themselves....

Support for non Tensor types for `ScriptModule`

2

#771 added some of types but not all. This GH issue is minor. HuggingFace has `scriptmodel` option in which the returned model only accepts ordinal arguments and outputs a tuple...

Implementation of `torchaudio.backend` missing

1

https://github.com/dotnet/TorchSharp/issues/805 We have stub functions `torchaudio.save()` and `torchaudio.load()`, however, we don't have implementation of https://github.com/dotnet/TorchSharp/blob/main/src/TorchAudio/AudioBackend.cs Similar issues with https://github.com/dotnet/TorchSharp/issues/471 TorchVision uses SkiaSharp to implement IO.