Soothsayer 0.5 released

Soothsayer 0.5 is now available for download.

Soothsayer 0.5 includes the new generalized smoothed n-gram statistical predictive plugin, which supports arbitrary order n-grams. Used in combination with the text2ngram tool, statistical predictions can be generated by n-gram language models of arbitrary cardinality. The new generalized smoothed n-gram predictive plugin also uses an improved heuristic to generate initial completion candidates, by using highest order n-gram statistics, and falling back on lower order n-grams if initial completion set is smaller than required.

This release also includes notable bug fixes and improvements to soothsayer simulator. A bug in the simulator caused the reported Key Stroke Reduction rate to much much lower than the actual KSR achieved by soothsayer.

Soothsayer 0.5 marks a change in project status from alpha to beta. However, it is still to be considered a developer’s preview release. This is a source release only. No precompiled packages or installers are provided.

Users wishing to try out Soothsayer will need to follow the (easy) steps required to build soothsayer on their machine, as detailed in the README file. Please note that SQLite is required to build soothsayer. CPPUnit is optional, but be aware that no unit tests will be built nor run when running `make check’, unless CPPUnit is installed.

Soothsayer 0.5 has been built and tested on various Linux platforms (including 64 bit architectures) and Windows XP + Cygwin platform. If you encounter any issues while building or running soothsayer, please report it to the author.

There is still a lot of work to be done. Currently, Soothsayer is training on a very small training corpus. Predictive performance can be greatly increased by using a larger training corpus. Users can easily generate statistical predictive resources using the text2ngram tool on a custom training text corpus.
New predictive plugins should also be added to take advantage of the multiple predictive source architecture.
Please refer to our TODO list for more details on what needs to be done.