Auto-captioning audio fails with 'NoneType' object has no attribute 'device' in low-VRAM mode

This is for bugs only

Did you already ask in the discord?

No - as previously noted, due to Discord's adherence to stupid UK age-verification laws I no longer use it due to privacy concerns.

You verified that this is a bug and not a feature request or question by asking in the discord?

Yes

Describe the bug

Attempting to auto-caption audio with low-VRAM enabled causes multiple 'NoneType' object has no attribute 'device' errors in the log, and no output is generated. (This was reported as #832 which got closed due to inactivity which seems to be common for bug reports here, but one of the comments contains a fix.)

Update: it fails whether low VRAM is enabled or not