Skip to content

Fix device mismatch in VibeVoiceASR.encode_speech under device_map=auto #441

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
Ammar Hashmi (ammar-iitm) wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Fix device mismatch in VibeVoiceASR.encode_speech under device_map=auto #441

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
5 changes: 5 additions & 0 deletions vibevoice/modular/modeling_vibevoice_asr.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -332,6 +332,11 @@ def _iter_segments(total_length: int, segment_length: int):

# Combine acoustic and semantic features
if speech_masks is not None:
# Under device_map="auto", acoustic_connector/semantic_connector can be
# dispatched to a different device than speech_masks (e.g. multi-GPU
# pipeline parallelism), and boolean-indexing requires the mask to be
# on the same device as the indexed tensor.
speech_masks = speech_masks.to(acoustic_features.device)
combined_features = acoustic_features[speech_masks] + semantic_features[speech_masks]
else:
combined_features = acoustic_features + semantic_features
Expand Down