Fix model conversion

[vietanhdev]

I've had this error on macOS and fixed it in this pr:

Downloading: "https:/dl.fbaipublicfiles.com/encodec/v0/encodec_24khz-d7cc33bc.th" to /Users/vietanhdev/.cache/torch/hub/checkpoints/encodec_24khz-d7cc33bc.th
Traceback (most recent call last):
  File "download_weights.py", line 41, in <module>
    state_dict = torch.hub.load_state_dict_from_url(
  File "/opt/homebrew/Caskroom/miniforge/base/envs/bark/lib/python3.8/site-packages/torch/hub.py", line 746, in load_state_dict_from_url
    download_url_to_file(url, cached_file, hash_prefix, progress=progress)
  File "/opt/homebrew/Caskroom/miniforge/base/envs/bark/lib/python3.8/site-packages/torch/hub.py", line 611, in download_url_to_file
    u = urlopen(req)
  File "/opt/homebrew/Caskroom/miniforge/base/envs/bark/lib/python3.8/urllib/request.py", line 222, in urlopen
    return opener.open(url, data, timeout)
  File "/opt/homebrew/Caskroom/miniforge/base/envs/bark/lib/python3.8/urllib/request.py", line 522, in open
    req = meth(req)
  File "/opt/homebrew/Caskroom/miniforge/base/envs/bark/lib/python3.8/urllib/request.py", line 1278, in do_request_
    raise URLError('no host given')
urllib.error.URLError: <urlopen error no host given>

For the running model conversion, I'm still missing vocab.txt and need to comment following part to convert the model successfully:

    # generate_vocab_file(vocab_path, out_dir / "ggml_vocab.bin")
    # print(" Vocab loaded.")

[PABannier]

Hi @vietanhdev ! Thanks for the PR! FYI, I've just pushed vocab.txt in ggml_weights. This is the vocab file from the bert-multilingual-cased tokenizer on HuggingFace.