Skip to content

Fix a compatibility issue with the timestamp format for Fun-ASR-Nano-2512 #2814

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
slin000111 wants to merge 4 commits into
base: main
Choose a base branch
from
Open

Fix a compatibility issue with the timestamp format for Fun-ASR-Nano-2512 #2814

Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
1fae103
fix prefix concatenation in timestamp post-processing for sensevoice
slin000111 Sep 17, 2025
e94ed5b
Remove unnecessary code
slin000111 Sep 17, 2025
76206c5
Merge branch 'modelscope:main' into main
slin000111 Feb 12, 2026
378a9ba
Fixed a compatibility issue with the timestamp format for fun-asr-nano.
slin000111 Feb 12, 2026
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: 6 additions & 2 deletions funasr/auto/auto_model.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -555,8 +555,12 @@ def inference_with_vad(self, input, input_len=None, **cfg):
if k not in result:
result[k] = []
for t in restored_data[j][k]:
t[0] += vadsegments[j][0]
t[1] += vadsegments[j][0]
if isinstance(t, dict):
t["start_time"] = (t["start_time"] * 1000 + vadsegments[j][0]) / 1000
t["end_time"] = (t["end_time"] * 1000 + vadsegments[j][0]) / 1000
Comment on lines +559 to +560
Copy link
Contributor

@gemini-code-assist gemini-code-assist bot Feb 12, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The calculation for adjusting timestamps can be simplified for better readability. Directly adding the offset in seconds is more concise and achieves the same result.

Suggested change
t["start_time"] = (t["start_time"] * 1000 + vadsegments[j][0]) / 1000
t["end_time"] = (t["end_time"] * 1000 + vadsegments[j][0]) / 1000
t["start_time"] += vadsegments[j][0] / 1000.0
t["end_time"] += vadsegments[j][0] / 1000.0

else:
t[0] += vadsegments[j][0]
t[1] += vadsegments[j][0]
result[k].extend(restored_data[j][k])
elif k == "spk_embedding":
if k not in result:
Expand Down