kaljurand / grammars Goto Github PK
View Code? Open in Web Editor NEWGrammatical Framework based speech recognition grammars for Estonian, English and potentially many more languages
Home Page: http://kaljurand.github.io/Grammars/
Grammatical Framework based speech recognition grammars for Estonian, English and potentially many more languages
Home Page: http://kaljurand.github.io/Grammars/
See various examples at:
The current solution of mapping placenames to themselves has several problems
There should be at least 2 linearization formats:
The downside is that the grammar gets 3 times bigger.
The Numeral-grammar currently supports numbers up do 999999 (10^6), but this is not enough for some common expressions, e.g. including some world currencies.
Large numbers are sometimes easier to express using fractions, e.g. it is more natural and shorter to say
kaks ja pool miljonit
or
kaks koma viis miljonit
than
kaks miljonit ja viis sada tuhat
as in
kui palju on kaks meetrit jalgades
Often this is more natural to say. Also it might reduce errors (e.g. when using the Action grammar which combines many grammars with similarly sounding sentences).
This script could be used in testing to make sure that no unwanted (or avoidable) approximation has occurred during the GF->JSGF conversion.
The script should highlight phrases which are accepted by the JSGF but not by the original GF grammar. We know that approximation happens during JSGF->FSA, but maybe it happens already during GF->JSGF in which case it can probably be avoided by redesigning the grammar.
Add support for mass <-> volume conversion parametrized by a closed set of substances (sugar, flour, coco powder, corn starch, and buckwheat) or a given density value 0.719 g/ml. E.g.
150 grammi jahu milliliitrites
150 grammi milliliitrites eeldusel et ühes liitris on 12345 grammi
Both Google and WolfraAlpha interpret:
-1^1 = -(1^1)
-1+1 = (-1) + 1
But we probably want
miinus X OP Y
to always stand for
(- X) OP Y
Example:
how much is two hours and thirty minutes in seconds
convert 2 h 30 min to s
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.