Sign in

username:

password:



Not a member?

Search speech-recognition



Search tips

Subscribe to speech-recognition



Discussion Groups

Discussion Groups | Speech Recognition | Non-learning Algorithm

Technical discussions about the implementation and research of speech recognition algorithms.

  

Post a new Thread

Non-learning Algorithm - Mounir Shita - Mar 7 16:11:00 2001



Hi everyone,

Do you guys know if there exist a ASR algorithm that doesn't require
learning and works for more than just one person?

For example, a gizmo you can turn on and instantly understands what
anyone have to say (or whatever the gizmo's vocabulary is).

Mounir





(You need to be a member of speech-recognition -- send a blank email to speech-recognition-subscribe@yahoogroups.com )

Re: Non-learning Algorithm - Richards - Mar 7 22:43:00 2001

Yeah, well, no. That is what we are all looking for!
/// JHR ///
__________________________________________________
----- Original Message -----

> Do you guys know if there exist a ASR algorithm that doesn't require
> learning and works for more than just one person?




(You need to be a member of speech-recognition -- send a blank email to speech-recognition-subscribe@yahoogroups.com )

Re: Non-learning Algorithm - Mounir Shita - Mar 8 0:25:00 2001

I see. I've been working on an algorithm for the last couple of years that doesn't require learning. However, I've only been able to get 70% recognition rate at the very best. Normally is lower than that.

However, I do have some ideas on how to improve it. But it will take me some time to do so since this is only a privat project of mine. Nothing to do with my job or anything.

Just asked since I really don't know much (read: nothing) about the speech recognition industry/news etc..etc...I got in to this by an accident. But I had my fun and didn't want to spend anymore time with it if it was already done. But since its not, it would be fun to see if I can make it better. I guess if I could get a min.80-90% recognition rate, I would be happy

Then again, this project might just be a stupid disaster. Never know.....

Mounir

>

>Yeah, well, no. That is what we are all looking for!
>/// JHR ///
>__________________________________________________
>----- Original Message -----
>
> > Do you guys know if there exist a ASR algorithm that doesn't require
> > learning and works for more than just one person?
>

Get your FREE download of MSN Explorer at http://explorer.msn.com





(You need to be a member of speech-recognition -- send a blank email to speech-recognition-subscribe@yahoogroups.com )