You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I'm testing the whisper-vq-stoks-medium-en+pl.model with RQBottleneckTransformer inside vq_stoks. I can encode an audio into stokens using encode_audio.
However, when I try to dequantize the stokens using dequantize, I am encountering the error:
---------------------------------------------------------------------------
TypeError Traceback (most recent call last)
Cell In[16], [line 1](vscode-notebook-cell:?execution_count=16&line=1)
----> [1](vscode-notebook-cell:?execution_count=16&line=1) vq_model.dequantize(stokens_female[0])
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/vq_stoks.py:517, in RQBottleneckTransformer.dequantize(self, stoks)
[515](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/vq_stoks.py:515) positions = torch.arange(0, x.shape[-2], dtype=torch.long, device=x.device)
[516](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/vq_stoks.py:516) x = x + self.positional_embedding(positions)
--> [517](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/vq_stoks.py:517) return self.ln_post(self.out_blocks(x))
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1553, in Module._wrapped_call_impl(self, *args, **kwargs)
[1551](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1551) return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
[1552](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1552) else:
-> [1553](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1553) return self._call_impl(*args, **kwargs)
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1562, in Module._call_impl(self, *args, **kwargs)
[1557](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1557) # If we don't have any hooks, we want to skip the rest of the logic in
[1558](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1558) # this function, and just call forward.
[1559](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1559) if not (self._backward_hooks or self._backward_pre_hooks or self._forward_hooks or self._forward_pre_hooks
[1560](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1560) or _global_backward_pre_hooks or _global_backward_hooks
[1561](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1561) or _global_forward_hooks or _global_forward_pre_hooks):
-> [1562](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1562) return forward_call(*args, **kwargs)
[1564](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1564) try:
[1565](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1565) result = None
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/container.py:219, in Sequential.forward(self, input)
[217](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/container.py:217) def forward(self, input):
[218](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/container.py:218) for module in self:
--> [219](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/container.py:219) input = module(input)
[220](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/container.py:220) return input
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1553, in Module._wrapped_call_impl(self, *args, **kwargs)
[1551](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1551) return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
[1552](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1552) else:
-> [1553](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1553) return self._call_impl(*args, **kwargs)
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1562, in Module._call_impl(self, *args, **kwargs)
[1557](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1557) # If we don't have any hooks, we want to skip the rest of the logic in
[1558](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1558) # this function, and just call forward.
[1559](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1559) if not (self._backward_hooks or self._backward_pre_hooks or self._forward_hooks or self._forward_pre_hooks
[1560](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1560) or _global_backward_pre_hooks or _global_backward_hooks
[1561](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1561) or _global_forward_hooks or _global_forward_pre_hooks):
-> [1562](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1562) return forward_call(*args, **kwargs)
[1564](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1564) try:
[1565](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1565) result = None
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:222, in ResidualAttentionBlock.forward(self, x, x_positions, xa, xa_positions, causal, mask)
[212](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:212) def forward(
[213](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:213) self,
[214](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:214) x: Tensor,
(...)
[219](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:219) mask=None,
[220](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:220) ):
[221](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:221) lnx = self.attn_ln(x)
--> [222](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:222) x = x + self.attn(lnx, x_positions, lnx, x_positions, causal=causal, mask=mask)
[223](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:223) if self.cross_attn:
[224](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:224) lnx = self.cross_attn_ln(x)
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1553, in Module._wrapped_call_impl(self, *args, **kwargs)
[1551](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1551) return self._compiled_call_impl(*args, **kwargs) # type: ignore[misc]
[1552](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1552) else:
-> [1553](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1553) return self._call_impl(*args, **kwargs)
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1562, in Module._call_impl(self, *args, **kwargs)
[1557](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1557) # If we don't have any hooks, we want to skip the rest of the logic in
[1558](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1558) # this function, and just call forward.
[1559](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1559) if not (self._backward_hooks or self._backward_pre_hooks or self._forward_hooks or self._forward_pre_hooks
[1560](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1560) or _global_backward_pre_hooks or _global_backward_hooks
[1561](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1561) or _global_forward_hooks or _global_forward_pre_hooks):
-> [1562](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1562) return forward_call(*args, **kwargs)
[1564](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1564) try:
[1565](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/torch/nn/modules/module.py:1565) result = None
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:129, in MultiHeadAttention.forward(self, qx, q_positions, kvx, kv_positions, causal, mask)
[126](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:126) q,k,v = None,None,None
[128](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:128) if q is None: q = self.query(qx) * self.sqrt_qk_scale
--> [129](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:129) q = self.split_heads(q, q_positions, rope = self.rotary, subsampling = self.query_subsampling)
[131](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:131) if kvx is not self.cached_kvx:
[132](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:132) if k is None: k = self.key(kvx) * self.sqrt_qk_scale
File ~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:108, in MultiHeadAttention.split_heads(self, x, x_positions, rope, subsampling)
[106](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:106) x = x.view(*x.shape[:2], self.n_head, -1)
[107](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:107) if rope:
--> [108](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:108) x = rope_rotate(x, x_positions * subsampling, *self.rotary(x))
[109](https://vscode-remote+attached-002dcontainer-002b7b22636f6e7461696e65724e616d65223a222f72657365617263685f6c6162227d-0040ssh-002dremote-002b7b22686f73744e616d65223a224a616e2d34784136303030227d.vscode-resource.vscode-cdn.net/home/phong/Workspace/WhisperVQ/~/Workspace/WhisperVQ/.venv/lib/python3.11/site-packages/whisperspeech/modules.py:109) return x.permute(0, 2, 1, 3)
TypeError: unsupported operand type(s) for *: 'NoneType' and 'int'
Looking for help, thank you in advance!
The text was updated successfully, but these errors were encountered:
Hi, thank you for the amazing works!
I'm testing the
whisper-vq-stoks-medium-en+pl.model
withRQBottleneckTransformer
insidevq_stoks
. I can encode an audio into stokens usingencode_audio
.However, when I try to dequantize the stokens using
dequantize
, I am encountering the error:Looking for help, thank you in advance!
The text was updated successfully, but these errors were encountered: