Skip to content

降低zero-shot功能对torch版本的要求 #638

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
fumiama merged 6 commits into from
Jul 28, 2024
Merged

降低zero-shot功能对torch版本的要求 #638

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
9e96be7
fix some problem
ZaymeShaw Jul 27, 2024
0820f3c
delete annotation
ZaymeShaw Jul 27, 2024
34e626a
Merge remote-tracking branch 'upstream/dev' into clone_test
ZaymeShaw Jul 28, 2024
a0c3e04
some simplify
ZaymeShaw Jul 28, 2024
0b7da4c
chore(format): run black on dev
github-actions[bot] Jul 28, 2024
f7a7e46
correct the device_gpt
ZaymeShaw Jul 28, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 1 addition & 0 deletions ChatTTS/core.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -297,6 +297,7 @@ def _load(
use_flash_attn=use_flash_attn,
use_vllm=use_vllm,
device=device,
device_gpt=self.device_gpt,
logger=self.logger,
).eval()
assert gpt_ckpt_path, "gpt_ckpt_path should not be None"
Expand Down
4 changes: 2 additions & 2 deletions ChatTTS/model/tokenizer.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -180,12 +180,12 @@ def _decode_prompt(prompt: str) -> torch.Tensor:
dtype="<u2",
).copy()
del dec
return torch.from_numpy(p).view(*shp)
return torch.from_numpy(p.astype(np.int32)).view(*shp)

@staticmethod
@torch.no_grad()
def _encode_prompt(prompt: torch.Tensor) -> str:
arr: np.ndarray = prompt.to(dtype=torch.uint16, device="cpu").numpy()
arr: np.ndarray = prompt.cpu().numpy().astype(np.uint16)
shp = arr.shape
assert len(shp) == 2, "prompt must be a 2D tensor"
s = b14.encode_to_string(
Expand Down