|
|
|
|
|
|
by mrtrombone
4912 days ago
|
|
|
Great links thanks. One thing I have been wondering is what text to speech synthesis would sound like with non 'human' voices. Most demo's are of course focussed on human voices e.g. http://www.cstr.ed.ac.uk/projects/festival/morevoices.html but it would be interesting to develop some 'creature' ones as per elmo, furby, alvin etc Have you heard the Karotz rabbit's voice? |
|
|
The less your animatronic creature does, the greater you project your own emotions onto it, and more likely you are to find it realistic or compelling or interesting or natural. Poor robotic speech instantly kills that empathy. Think about R2-D2's nonsensical beeps (that express emotion tonally) versus C3-PO's obnoxious prattling.
If you're trying to make it be "helpful" by telling you status information, how many times have you been frustrated when someone gives you directions, or is telling you about a problem narratively instead of in terms of reproducible steps? Speech is very inefficient, but for empathizing, Furbish was great.