Is there a way around this, or perhaps a 3rd party voice recognition app that gets around it?
I also ran a few tests in a Scouse accent and it didn't like any of them - are people with strong regional dialects ever going to be able to use this technology?
Is there a way of training it to better recognaise the nuances of your voice?
In the past this sort of technology works best with:-
- no training but a low number of words
- high amount of training allowing a larger recognaised vocabulary
It's fair to assume the technology has moved on a lot but there's still scope for it to be trained to an extent. We'd all prefer it to work out of the box of course.