Deprecated: mysql_connect(): The mysql extension is deprecated and will be removed in the future: use mysqli or PDO instead in /home/virtualw/public_html/Resources/Hosted/Resource.php on line 9
Advanced Speech Encoding
Not a member yet? Register for full benefits!

Username
Password
Advanced Speech Encoding

One of the many plans which does show promise for a virtual voice - that is, a voice which has no real correspondence to the physical voice, or lack thereof, is a system called Advanced Speech Encoding which DARPA (the Defence Advanced Research Projects Agency) is working on.

Advanced Speech Encoding, or ASE which is designed to reduce the number of bits required to transmit voice signals over a data stream down to the minimum possible. It is designed for burst transmissions, but has applications in many more areas. The immediate impact of the technology is very obvious, as it works to allow voice communication with very minimal data requirements, allowing much more data to be packed in with it, than traditional voice signals allow. For virtual environments, this would be ideal, as voice traffic takes up huge amounts of packet space, radically reducing the rest of the data flow.

The ASE program's goal is to have recognisable speech transmission in as little as 300 bytes a second, which is less than one twentieth of normal. This of course, reduces it almost to random noise; normally it would be impossible to pick up any usable information at this level. However, the voice signals produced will still be machine-recognisable, and understandable by voice recognition software.

This is the key point of interest to virtual reality systems, as once the thin transmission reaches voice recognition software, the speech has already been reduced to the lowest common denominators, with perhaps the barest inflection remaining. Thus it should be possible to re-augment the speech by other technologies, which would have a common base to work from to transform your own speech into that of an orc, dragon, or sentient computer.

The technology is still in very infantile stages, and a great deal of work lays ahead in re-converting the minimised speech into something with the desired inflection and tonal range. However, if successful, this technology offers the ability to radically downsize the communication overhead on a collaborative VR system.

Links

Defence Advanced Research Projects Agency

ASE

Staff Comments

 


.
Untitled Document .