On 3/6/2013 4:18 PM, meekerdb wrote:
On 3/6/2013 10:45 AM, John Clark wrote:
"I just saw the movie "Beowulf", it's a pretty good movie but what is of interest is the stunning advance in animation achieved by good old Moore's Law. There were times when I could swear I was looking at a real human being not something a computer produced. However there is something puzzling, all the voices in the movie were still made by human beings. An innocent might think that as video is a much higher bandwidth media than audio video would be harder to simulate than audio, but apparently that is not the case."

I don't think it's that audio is harder to synthesize, I think it's just that audio is a lot cheaper to produce the old fashioned way. If you wanted some really unusual audio special effects it might be cheaper to synthesize them.

Hi Brent,


I have it from very good authority (AT&T/Bell Labs folks) that authentic sounding speech synthesis is "a bitch". The company that I am working for is laboring long and hard to deal with this problem. This difficulty, BTW, makes voice recognition systems hard to implement for authentication purposes. The fact that humans can do it very well is very telling of the ingenuity of evolutionary systems.



You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to everything-list+unsubscr...@googlegroups.com.
To post to this group, send email to everything-list@googlegroups.com.
Visit this group at http://groups.google.com/group/everything-list?hl=en.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to