I hope this isn't a big job but I would be willing to pay. I don't really have much but if it is worth $100 for someone to do this, I would be happy to pay via paypal. My reason for asking this is that I think it will take me a really long time to figure this out as I am farily new to this and have some specific deadlines for my project.
I need 2 utility classes (in Java):
1) Class that has a method which accepts a URL or File which will return the String evaluation result using the standard US alphabet (will be used for a person's name)
2) Class that has a method which accepts a URL or File which will return the String evaluation result using the standard US alphabet, special characters, and numeric digits (will be used for email addresses)
The File input will be encoded WAV files created by Asterisk.
I could provide some sample WAV files for testing purposes and all you would have to do is create a JUnit test which demonstrates the evaluation.
Would anyone be interested in doing this? Thanks.
Regards,
Joe Hudson
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
> Class that has a method which accepts a URL or File which will return the String evaluation result using the standard US alphabet (will be used for a person's name)
Both are very complicated tasks btw even if the list of names is fixed. If not and you want to recognize arbitrary name, the complexity goes to almost impossible :)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Well I doubt you will find someone to do that for you but if your willing to do it yourself, check out http://scribblej.com/svn/
It integrates sphinx with asterisk.
Or as it sounds like you might be interested in Media Resource Control Protocol. MRCP abstracts out the speech recognition from asterisk. So the number crunching doesnt take place on the asterisk server but a dedicated speech server. Take a look at cairo at http://www.speechforge.org/
Its actually easy to get the cairo demos going with the default acoustic/language models.
Getting a mediocre system will take no time at all, probably 2 hours as cairo has some good demos. Fine tuning it will take time. Lots.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I hope this isn't a big job but I would be willing to pay. I don't really have much but if it is worth $100 for someone to do this, I would be happy to pay via paypal. My reason for asking this is that I think it will take me a really long time to figure this out as I am farily new to this and have some specific deadlines for my project.
I need 2 utility classes (in Java):
1) Class that has a method which accepts a URL or File which will return the String evaluation result using the standard US alphabet (will be used for a person's name)
2) Class that has a method which accepts a URL or File which will return the String evaluation result using the standard US alphabet, special characters, and numeric digits (will be used for email addresses)
The File input will be encoded WAV files created by Asterisk.
I could provide some sample WAV files for testing purposes and all you would have to do is create a JUnit test which demonstrates the evaluation.
Would anyone be interested in doing this? Thanks.
Regards,
Joe Hudson
> Class that has a method which accepts a URL or File which will return the String evaluation result using the standard US alphabet (will be used for a person's name)
Both are very complicated tasks btw even if the list of names is fixed. If not and you want to recognize arbitrary name, the complexity goes to almost impossible :)
Well I doubt you will find someone to do that for you but if your willing to do it yourself, check out
http://scribblej.com/svn/
It integrates sphinx with asterisk.
Or as it sounds like you might be interested in Media Resource Control Protocol. MRCP abstracts out the speech recognition from asterisk. So the number crunching doesnt take place on the asterisk server but a dedicated speech server. Take a look at cairo at http://www.speechforge.org/
Its actually easy to get the cairo demos going with the default acoustic/language models.
Getting a mediocre system will take no time at all, probably 2 hours as cairo has some good demos. Fine tuning it will take time. Lots.