Skip to content

fix: resolve tensor shape mismatch and index out-of-bounds in CausalM… #1848

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

Open
Caxson wants to merge 1 commit into
base: main
Choose a base branch
from
Open

fix: resolve tensor shape mismatch and index out-of-bounds in CausalM… #1848

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
8 changes: 7 additions & 1 deletion cosyvoice/flow/flow.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -347,12 +347,18 @@ def forward(
h = h.repeat_interleave(self.token_mel_ratio, dim=1)
mask = mask.repeat_interleave(self.token_mel_ratio, dim=1).squeeze(dim=-1)

# align feat and h lengths: speech_token (parquet) and speech_feat (pipeline) may differ
# due to different alignment (e.g. 960 padding), use min length to avoid tensor size mismatch
min_len = min(feat.shape[1], h.shape[1])
feat = feat[:, :min_len, :]
h = h[:, :min_len, :]
mask = mask[:, :min_len]
# get conditions
conds = torch.zeros(feat.shape, device=token.device)
for i, j in enumerate(feat_len):
if random.random() < 0.5:
continue
index = random.randint(0, int(0.3 * j))
index = min(int(0.3 * j), min_len)
conds[i, :index] = feat[i, :index]
conds = conds.transpose(1, 2)

Expand Down