Closed a897456 closed 4 months ago
The output speech after going through the codec and the reference speech do have differences in duration or sampling points, which is inevitable. This should not be an obstacle for using MUSHRA. Is there any way to solve this?
### Tasks
I didn't read the experiment manual. It's solved. Sorry.
The output speech after going through the codec and the reference speech do have differences in duration or sampling points, which is inevitable. This should not be an obstacle for using MUSHRA. Is there any way to solve this?