Xavier Anguera

Hi, I would like to use the finetuned Hubert models available from [https://github.com/facebookresearch/fairseq/tree/main/examples/hubert]() When I try to load them with, for example `model_3 = getattr(hub, 'hubert_url')('https://dl.fbaipublicfiles.com/hubert/hubert_large_ll60k_finetune_ls960.pt')` I am getting the...

Alignment outputs are not as expected

5

Hi, I am using phonetisaurus to align a a grapheme input to its phonetic transcription. For this I use the phonetisaurus-align tool with alignment models trained on CMUDict. I a...

unable to apply voice conversion to long files using my trained speaker embedding

4

Hi, I am attempting zero-shot voice conversion by using only a few audio sentences of a target speaker. I am training a speaker embedding for this speaker using the make_spect.py...

Is there a way to respond to ping messages?

4

Hi, I am trying to setup a connection between Tubesock and the Android library okhttp-ws. It looks like okhttp-ws sends pings but does not provide ping response mechanism. I see...

Hi, I would be grateful if you could consider adding a license to the code. Apache or MIT would be great, so that it makes it most flexible to be...

What are the GPU memory specs needed to run pretraining and kmeans?

1

Hi, I am trying to run the pretraining of the full model (which should have ~650M parameters) in a 24GB GPU card and it only runs if I set the...

After calculating embeddings I get an error about a worker lost, with no more info on what could be the cause. Any idea how to solve it? ``` worker-1 |...