Andrew Hunt wrote: > One approach is that the score > reflects the "acoustic match" [...] > Another approach tries to > determine how likely the guess is to be correct [...] > There are many other variations > and many bright people have spent good time debating the relative merits > If there are multiple different notions of confidence, shouldn't there be multiple different "confidence" scores in VoiceXML? Is there at least some core set of confidence measures that one could settle on and make into different "confidence" properties?
Received on Friday, 2 November 2007 06:38:51 UTC