This can happen when you select "Mic + Computer" as the audio input on Caption.Ed, but then play your computer audio out of your speakers without using headphones.
When this happens, Caption.Ed effectively tries to process your computer audio twice with a slight delay; once internally and then a second time when your microphone picks up the audio coming from your speakers.
This echoing effect has a massive impact on the accuracy of the automatic captions.
To avoid this, we recommend using headphones if possible.
Using headphones will mean that Caption.Ed can still capture the internal audio the first time without it looping back through your speakers and microphone.