[GH-ISSUE #1305] High "Extra inference time" value with realtime GUI leads to quieted/cutoff words #846

New issue

Open

opened 2026-10-20 10:03:35 +00:00 by bsakaguchi · 1 comment

bsakaguchi commented

2026-10-20 10:03:35 +00:00

Owner

Originally created by @Mojobones on GitHub (Sep 21, 2023).
Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/1305

I notice that if I set the extra inference time to a high value (most prominent at 5, but starts to appear above ~1.5) that the first words I speak will be extra quiet, if not cut out entirely. This only happens if there's a long pause between speaking (ex. waiting 7s between spoken words consistently reproduces it.) At a value of 1.5 or lower, this issue does not happen and there's no issues hearing the initial words i speak.

Is there any insight into what could be causing this? It happens across 32/40/48k models and various build parameters. @yxlllc sorry for the ping but you seem to engage with the realtime usage a lot, and i'm curious if you can reproduce this behavior.

Edit: For added context, I trim all silences out of my build audio, and have confirmed that others can reproduce this problem. Interestingly, it's the least pronounced on 32k models.

Originally created by @Mojobones on GitHub (Sep 21, 2023). Original GitHub issue: https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI/issues/1305 I notice that if I set the extra inference time to a high value (most prominent at 5, but starts to appear above ~1.5) that the first words I speak will be extra quiet, if not cut out entirely. This only happens if there's a long pause between speaking (ex. waiting 7s between spoken words consistently reproduces it.) At a value of 1.5 or lower, this issue does not happen and there's no issues hearing the initial words i speak. Is there any insight into what could be causing this? It happens across 32/40/48k models and various build parameters. @yxlllc sorry for the ping but you seem to engage with the realtime usage a lot, and i'm curious if you can reproduce this behavior. Edit: For added context, I trim all silences out of my build audio, and have confirmed that others can reproduce this problem. Interestingly, it's the least pronounced on 32k models.

bsakaguchi added the

following up

label

2026-10-20 10:03:35 +00:00

bsakaguchi commented

2026-10-20 10:03:36 +00:00

Author

Owner

@yxlllc commented on GitHub (Oct 7, 2023):

This may be due to an error in contentvec inferring silence, we may try an improved approach in the future.

@yxlllc commented on GitHub (Oct 7, 2023): This may be due to an error in contentvec inferring silence, we may try an improved approach in the future.