Cannot succeed inference from linked colab notebook #91
Description
I was trying to run your linked colab example on T4 GPU. However, The following cell getting error:
deep_clone = True # set to False if you don't know prompt transcript or want fast inference.
# Below you can tune other inference settings, like top_k, temperature, top_p, etc...
cfg = config_class(deep_clone=deep_clone, rep_penalty_window=100,
top_k=100, temperature=0.7, freq_penalty=3)
ar_codes, wav_out = mars5.tts("The quick brown rat.", wav,
ref_transcript,
cfg=cfg)
print('Synthesized output audio:')
ipd.Audio(wav_out.numpy(), rate=mars5.sr)
The error is:
Note: using deep clone. Assuming input `c_phones` is concatenated prompt and output phones. Also assuming no padded indices in `c_codes`.
New x: torch.Size([1, 636, 8]) | new x_known: torch.Size([1, 636, 8]) . Base prompt: torch.Size([1, 215, 8]). New padding mask: torch.Size([1, 636]) | m shape: torch.Size([1, 636, 8])
---------------------------------------------------------------------------
ValueError Traceback (most recent call last)
[<ipython-input-8-2d05018561f0>](https://localhost:8080/#) in <cell line: 0>()
4 top_k=100, temperature=0.7, freq_penalty=3)
5
----> 6 ar_codes, wav_out = mars5.tts("The quick brown rat.", wav,
7 ref_transcript,
8 cfg=cfg)
6 frames
[~/.cache/torch/hub/Camb-ai_mars5-tts_master/mars5/trim.py](https://localhost:8080/#) in as_strided(x, shape, strides, subok, writeable)
544 """
545 # first convert input to array, possibly keeping subclass
--> 546 x = np.array(x, copy=False, subok=subok)
547 interface = dict(x.__array_interface__)
548 if shape is not None:
ValueError: Unable to avoid copy while creating an array as requested.
If using `np.array(obj, copy=False)` replace it with `np.asarray(obj)` to allow a copy when needed (no behavior change in NumPy 1.x).
For more details, see https://numpy.org/devdocs/numpy_2_0_migration_guide.html#adapting-to-changes-in-the-copy-keyword.
Is it possible to update the colab notebook with the fix?
Thanks in advance