deepgeek/CosyVoice
2
0
Fork 0
mirror of https://github.com/FunAudioLLM/CosyVoice.git synced 2026-02-05 18:09:24 +08:00
Code Issues Packages Projects Releases Wiki Activity

Refactor CUDA stream context management in CosyVoice2Model

Browse Source 
- Replaced the use of torch.cuda.stream with a direct context management approach for improved clarity and performance during inference.
- This change simplifies the stream handling code while maintaining efficient resource utilization.
This commit is contained in:
禾息
2025-04-16 16:04:40 +08:00
parent 369f3c2c18
commit e8a26827ae
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
cosyvoice/cli/model.py
View File

@@ -385,7 +385,7 @@ class CosyVoice2Model(CosyVoiceModel):
# this_uuid is used to track variables related to this inference thread # this_uuid is used to track variables related to this inference thread
self.synchronize_stream() self.synchronize_stream()
stream_context = self.stream_context_pool.get() stream_context = self.stream_context_pool.get()
with torch.cuda.stream(stream_context): with stream_context:
this_uuid = str(uuid.uuid1()) this_uuid = str(uuid.uuid1())
with self.lock: with self.lock: