Ok, I get a statistic about which sounds I most often get wrong but there is no targeted pratice to improve then. For the report it is totally unclear how the well spoken percentage is calculated, is it everything ever recorded, the last week, nothing tells you.