Reducing false accepts with decoys
As discussed in a previous post, one of the unfortunate consequences of out-of-grammar utterances is that they can cause many false accepts that may seriously degrade application performance and user...
View ArticleComparing different speech recognition engines
We’re sometimes asked to compare the performance of different speech recognition engines on an identical task (same grammar, same set of test utterances). To do so in an effective way, we rely on three...
View ArticleWhat’s the point of having 98% in-grammar accuracy if 40% of user utterances...
How many times have you heard people say that they “achieve 95% speech recognition accuracy” (or more)? That sounds really impressive, doesn’t it? It shouldn’t. What they don’t tell you is that they...
View ArticleEffective sentence generation
Following some interesting discussions on the Yahoo VUIDs group about sentence generation, I give some thoughts on this subject and present the tools available in NuGram Pro (video included). [...]
View ArticleGrammar problem #1 – repeated tokens
It is quite easy to write a speech recognition grammar. After all, it’s only a text file. And with the help of a good editor, we can expect the grammar to be free of syntax errors, i.e. to conform to...
View ArticleClik here to view.
Nu Echo presents SpeDial project results to the European Commission;...
On January 14th, I will be in Luxembourg to present the results of a joint research project to reviewers of the European Commission.We’ve not been vocal about it (and frankly, I’m not looking for...
View ArticleClik here to view.
SpeDial project was a great success!
The SpeDial partners. From left to right: Dominique Boucher (Nu Echo), Fernando Batista (INESC-ID), Katerina Louka (VoiceWeb), Isabel Trancoso (INESC-ID), Joakim Gustafson (KTH), and coordinator...
View Article
