Hi folks,
I would like to ask your help with an open-source project. Ivela-Voice. An applet (written in Java 1.6) of Voice Recognition, that aims to be a part of an E-learning application, for an Open English Course.
I need people to access the test page of project and try to use this "beta applet".
You will be prompted with 10 words in english, then you will need to repeat these same words, finally the application will try to recognize what you said.
This test is an anonymous process. It will take no more than 5 minutes. The test results will be submited to a database, for a statistic analysis.
In order to get this software working, you will need to access through Firefox >= 2.0 with Java 1.6 plugin installed. The Operating Systems supported are Linux and Windows.
The url to test is:
http://200.17.41.215:8080/ivela-web/ivela_voice_stat.html
Follow the instructions of test page. This is a good opportunity to contribute with Open-Source Community :-)
May the Force be with us all.
Jefferson de Carvalho
--- (Edited on 10/28/2008 3:43 pm [GMT-0500] by jeffersoncarvalho) ---
Tested with Firefox 3 under windows, the page gives me error 404.
type Status report
message
descriptionThe requested resource () is not available.
Only for you to know.
Regards,
Ivan
--- (Edited on 11/3/2008 6:42 am [GMT-0600] by ubanov) ---
Hi Jefferson,
Works with FireFox 2.0 on Windows Vista, Java standard edition 6, update 4 (build 1.6).
Very cool!
What are you doing with the submitted audio? Will you be making it available to other projects?
When will you be making the code available?
thanks,
Ken
--- (Edited on 11/3/2008 9:30 pm [GMT-0500] by kmaclean) ---
hi Ken,
First of all, thank so much for your reply. Let's answer you questions.
Actually we are not submitting the audio file. There is no recording routine in our applet. I will explain what really happens:
when you load the page, my applet will search for the lastest version of julius under c:\julius (hidden folder in windows) or <your_home>/.julius (linux case). If there is no julian neither its grammars for our tests in you Hard Drive, my applet will download the suitable zip archive for each operational system.
After the download, the zip archive is unpacked and julian is ready to run. This happens only for the first time you load the page. In conclusion, it is a FAT client running julius. That's why th applet is signed.
Our project will be of course open source with GPL free license. The applet code is not done yet and needs improvement. After that I will make it able to comunity. This version you see is an "ugly" version. We have a much better that we will release as soon as possible. We'll do same to the audio files we have in our database.
Regards,
Jefferson
--- (Edited on 11/4/2008 12:09 pm [GMT-0600] by Visitor) ---
Awesome work!
Does it also send statistical analysis of the speech (to build an acoustic model) or only error rates of the person's speech recognized by the applet?
--- (Edited on 11/5/2008 4:45 am [GMT-0600] by Visitor) ---
Hi Visitor,
Thanks, I appreciated your opinion. Well, our applet does not save any statical analysis. It only evaluates a person speech.
--- (Edited on 11/5/2008 9:20 am [GMT-0600] by Visitor) ---
Thanks, it was me. I look forward to the GPL-source from the better applet :)
--- (Edited on 05-11-2008 10:05 pm [GMT+0100] by dano) ---
Ok Dano, it will be available as soon as possible.
--- (Edited on 11/6/2008 5:44 am [GMT-0600] by jeffersoncarvalho) ---