export_to_ketos_protobuf imcompatible with load_audio_representation() with the name parameter
Basically, when trying to use the export to protobuf function, it expects the audio_representation to be in the following format:
{'name': audio_representation}
However when loading the representation with load_audio_representation(audio_config_file, name="name") we get just the audio_representation object, which breaks the following line of code in the export function:
for ar in audio_repr.values():
assert 'step' in ar.keys()Even if we have step in the representation, the wrong values are being looped, triggering the assert.
Is this an oversight on our part or intended? If so, I think we need to make clear not to use the name parameter when loading the audio representation
To upload designs, you'll need to enable LFS and have an admin enable hashed storage. More information