bedahr - Slashdot User

Comment Re:Open Source, or Microsoft-Owned? (Score 1) 140

by bedahr on Tuesday January 22, 2008 @07:58AM (#22136674) Attached to: Open Source Speech Recognition

You would only need to change the modelmanager (modelmanager.cpp) to use different tools.

It wouldn't be that hard I guess....

-- bedahr

Open Source Speech Recognition 140

Posted by CmdrTaco on Saturday January 19, 2008 @11:14AM from the hello-computer dept.

bedahr writes "The first version of the open source speech recognition suite simon was released. It uses the Julius large vocabulary continuous speech recognition to do the actual recognition and the HTK toolkit to maintain the language model. These components are united under an easy-to-use graphical user interface. Simon can import dictionaries directly from wiktionary (a subproject of wikipedia) or from files formated in the HADIFIX- or HTK format and grammar structures directly from personal texts. It also provides means to train the language model with new samples and add new words."

Submission + - Open Source speech recognition suite goes alpha (sourceforge.net)

Submitted by

bedahr

on Friday January 18, 2008 @12:11PM

bedahr writes: "The first version of the open source speech recognition suite simon was released.

It uses the Julius large vocabulary continuous speech recognition to do the actual recognition and the HTK toolkit to maintain the language model.

These components are united under an easy-to-use graphical user interface.

Simon can import dictionarys directly from wiktionary (a subproject of wikipedia) or from files formated in the HADIFIX- or HTK format and grammar structures directly from personal texts.
It also provides means to train the language model with new samples and add new words.
With the recognition results it can execute programs, type texts and open places.

Simon "encapsulates" the julius engine in the seperate "juliusd" program and communicates with it over TCP/IP (so that the actual recognition could even be done on a central server).

This opens a whole new door for open source speech recognition which was really limited to "tech-demos" (the julius enginge only has a commando line interface which just outputs the recognition results) and is the first step to a serious competition for dragon-naturally-speaking-under-wine which was the only (hardly functioning and commercial) way to get functional dictation under linux.

The target language is primarly German but the ui has been translated to English and the software can handle language models independant of their language (julius for example was primarly developed for Japanese).

As the currently released version is only 0.1-alpha-1 it is indeed unstable and feature-incomplete but the basic system is working and it looks very promising."

Slashdot Top Deals