Simple Parser for Natural Language?

Bob Ingria ingria at world.std.com
Sun Jul 25 21:07:03 UTC 1999


At 01:09 AM 7/17/99 +0400, Luciano Notarfrancesco wrote:
>
>Dan, I'm glad to hear you want to do this...
>If you just add speech synthesis and recognition... and a red eye...
>
>I have been thinking about doing natural language processing in Squeak, but I got the conclusion that the most natural way to communication with Squeak is the Smalltalk language already implemented. Anyway, doing natural language understanding even in a limited form would be really great! And if you will talk about Squeak objects, then you don't need any knoweledge base but the image it self!

Yes, but you would also have to implement a #describe message on Object, which would answer the requisite NL description.  The implementation of this method is non-trivial.  (And too large for the margins of this message...)

>I was planning to implement NGrammars for doing syntactic analysis in order to do better prosody generation for a text-to-speech system. I believe the N-grams stuff could be useful for your project.

I've seen statistical techniques used for NL analysis, but never N-grams, because they're not really suited for understanding/analysis.  The typical application in a speech recognition program is to filter the recognizer's hypotheses.  They can be very useful in that application.  Tri-gram filtering usually leaves in a lot of amusingly non-English schmutz.  But four-gram filtering produces utterances which are all valid English utterances.


-30-
Bob Ingria
As always, at a slight angle to the universe





More information about the Squeak-dev mailing list