©2003 Kåre Sjölander and Jonas Beskow
Introduction Download Documentation Links Forum
WaveSurfer user forum

FAQ  FAQ
Search  Search
Memberlist  Memberlist
Register  Register
Profile  Profile
Login  Login

URGENT: WaveSurfer Inquiries

 
Post new topic   Reply to topic    www.speech.kth.se/wavesurfer Forum Index -> WaveSurfer
View previous topic :: View next topic  
Author Message
janice_2k



Joined: 01 Dec 2004
Posts: 1

PostPosted: Wed Dec 01, 2004 8:30 am    Post subject: URGENT: WaveSurfer Inquiries Reply with quote

Dear Sir/Mdm,

I am a graduate student doing project related in speech recognition for e-education purposes. The project is from the Ministry of Education. I happened to see WaveSurfer from your website and would like to know more about it, to see if it can assist me in accomplishing the project. I wonder if anyone could assist and guide me in completing this project in time. I am very new in this area and desperately looking for assistance and help from specialists like your group. I donít know who should I refer to, therefore, it would be great if someone here could lend me a helping hand.

This project is for a business purpose and I am given a very short time to complete everything (2 months). I am afraid that I won't have enough time to do the research on my own, so I need a solution that can be easily incorporated into my teamís product. Basically we have to do a web-based English (British English) courseware for secondary school students. One section of the courseware is called Pronunciation. The courseware has to be SCORM-compliant (the courseware has to comply with Shareable Content Object Reference Model. It's a standard that all e-learning companies should be aware of. By following the standards, any courseware can be shared with other e-learning providers with no difficulty) and will be uploaded onto the server at each school. Students will be able to browse the courseware at their schools. The end product should be cross-platform, so it should run on Windows, Mac and Linux. Below is a description of how we are required to do for the courseware.

1. When a word is displayed, the user will have to record his voice. Once recorded, the system will evaluate whether the user's recorded pronunciation is correct and will give appropriate feedbacks. The user will also be able to click on a button to listen to the model pronunciation.

2. Another requirement is that the user will be able to see the wave spectrum of his recorded audio. The wave spectrum will be displayed next to the wave spectrum of the model answer.

We mainly use Macromedia Flash to develop the courseware. The output will be Flash movies (.swf files) embedded in HTML files. We are looking for a solution that can be easily integrated with the Flash-based courseware. We did find an ActiveX control that can do similar but not the same job. We have to pay a certain amount for the product for development purposes. In addition, we have to register our domain name and pay an annual fee in order to use the CAB file. For example, if we host the courseware at www.aaa.com we have to register the domain name. If we have a mirror site at www.bbb.com, we have to separately register and pay for that domain.

It's actually not a good solution since each school will have its own server, so students will be accessing the courseware from their local servers and all the domain names have to be registered with the company that produces the ActiveX control. We're looking for something that won't cause extra burden to the students, something that requires minimal download and installation procedure.

I noticed that WaveSurfer can meet parts of our project requirements, including multi- platform and customizable. It should be something close to what we are looking for in my 2nd requirement above (please correct me if I am wrong). I know a little of C but know nothing about Snack ToolKit and Tcl/Tk / Python. I wonder if someone could assist me in customizing the WaveSurfer? Basically, I don't need so many buttons. I just need 4-5 buttons, which includes "Play", "Stop", "Record", "Compare" and "Save". Besides, I just need waveform. The others like spectogram etc are not required. Is there a way to customize WaveSurfer is such a way that it can compare/analyse the user's recorded audio against the model audio (we, developer prerecord the standard pronunciation)? Can the WaveSurfer be publicly distributed, as we are planning to use on a webpage to be uploaded on the internet. I notice that the WaveSurfer is in an exe file format. Will this limit it to windows independent? As far as I know, exe file cannot be run on cross platform (correct me if I am wrong).

I have search for resources and helps extensively for the past couple of weeks but we still canít find anyone or organization that can lend us a helping hand. Most of them said our requirements canít be accomplished. Our company is quite keen in outsourcing this part of the project to some experts like you, if our available funds allow. How much is the cost for developing such a web-based application?

I would like to know if anyone has ever done any R&D similar to the description above. Do you know anyone or organization that can assist us in this area? We are looking for experts in developing and achieving our project in 2 months time. So I really hope if someone could provide some suggestions as soon as possible as I am looking for a possible solution very urgently. Looking forward to your fast reply and have a pleasant day.


Thanks in advance,
_________________
Janice
Back to top
View users profile Send private message MSN Messenger
kare



Joined: 16 Aug 2002
Posts: 371
Location: KTH, Centre for Speech Technology

PostPosted: Wed Dec 01, 2004 6:07 pm    Post subject: Reply with quote

I've actually created a demonstration application that seems to fit your description very closely. It used WaveSurfer components for visualization and our in-house version of Snack for the ASR based utterance comparison.
WaveSurfer's license is very liberal there are no problems sellings products based on it. Regarding the platform issue, the solution is probably to have three different binaries to download, one for each platform.
There are of course issues, biggest one being the performance of the pronunciation scoring. There are many ways of doing that, depending on what you think is important. Also, which languages should the system be able to handle?
Please send me a direct e-mail if you want more info, kare@speech.kth.se

KŚre
Back to top
View users profile Send private message Visit posters website
Display posts from previous:   
Post new topic   Reply to topic    www.speech.kth.se/wavesurfer Forum Index -> WaveSurfer All times are GMT + 1 Hour
Page 1 of 1

 
Jump to:  
You can post new topics in this forum
You can reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB 2.0.1 © 2001, 2002 phpBB Group