← Back to team overview

voice-driven-ui team mailing list archive

To Developers: Welcome to Voice Driven UI (SpeechControl)

 

*********************************************************
***   This message is directed for the developers.    ***
*********************************************************

Due to many "inquires" as to what language we're using, and what are the available projects to be worked on; I've devised the following as a blueprint for us for now.
I'm in the process of constructing Launchpad project pages and branches that you can push code to for easy access and more immediate examination.
I've just learned how to use Bazaar + Launchpad, and it seems to be the standard for Launchpad. If another version control mechanism is to be suggested, please reply with a description on how to use it.

Projects At Hand:
 (*) python-openmary
 (*) openmary-c++
 (*) openmary-c++-cli

Q: WHAT IS openMary?
A: openMary (http://mary.dfki.de/) is a speech synthesis application that's keen to our project for the fact that it can render "emotional" speech, in better terms, speech that sounds more aesthetically pleasing to the end user. We want to use this application in our project and permit other projects to use it as well. However, openMary is written in Java, and runs as a local HTTP server on the client's computer. We can communicate with the hosted version, but that's some extreme bandwidth. Our goal is to create libraries in our target programming languages for this application so that its power can be harnessed elsewhere.

Q: WHAT LANGUAGES should we code in?
A: Any language of the following that you feel comfortable with; either Python or C++. These two languages are indisputable (to some) for their speed and power, so we'd appease to the masses in terms of library making.

Q: WHAT LICENSE should I code my application under?
A: For now, I'm assuming that the GPL2 license is good enough. Licensees aren't a developer issue, at least not at the moment.

Q: HOW CAN I track the progress of the other projects?
A: For now, *this* mailing list is the primary means of being updated until the Launchpad project pages are set up by either myself (JackyAlcine) or another **appointed** developer.

Q: WHAT should I do now?
A: If you can code, then you should take a look below.

The C++ and Python libraries (or scripts) have to be able to do the following:

	- Provide a *simple* means of initializing, utilizing, and ending a openMary interaction session.
	  An openMary interaction session doesn't consist of only **one** processed chunk of data; it consists of a whole session of processed and received data.

	- Expose the API shown here (http://mary.dfki.de:59125/documentation.html); it's included as a PDF and HTML document to this email.

	- The approach of doing so should conform to an object-orientated style of programming; to ensure software reuse.

NOTE: You might need a bit of understanding of XML schemas and SSML for you to fully encapsulate the openMary web API.

Thank you for reading this e-mail (typed it up for 20 mins =]).
Have a good day, evening, night, whatever. 
-- 
Jacky Alcine <jackyalcine@xxxxxxxxx>
Launchpad: https://launchpad.net/~jackyalcine
Wiki: https://wiki.ubuntu.com/jackyalcine
Blog: http://jdevelopthis.blogspot.com

Attachment: OpenMary HTTP Interface Documentation.pdf
Description: Adobe PDF document

Attachment: signature.asc
Description: This is a digitally signed message part