TIP: Click on subject to list as thread! ANSI
echo: electronics
to: MIKE ROSS
from: JAY EMRIE
date: 2002-12-13 22:45:00
subject: TALKING BOOKS

MR> MR> OTOH speech to text is actually quite easy

MR>


MR> JE> I wonder why, if it is so easy. is it so hard for voice
MR> JE> recognition programs to function reasonably accurately???

MR>Oh, rats! Yet another case of forgetting to proofread what goes out the
MR>mail box. Thanks for pointing it out! Indeed it is "text to
speech" and
MR>not the other way around.

MR>Mind you that I am definitely impressed with the speech recognition
MR>program in the phone system at the airport. It never has goofed each
MR>time I've had to use it. Then again it is usually simple numbers so I
MR>suppose the program already knows what to expect.

MR>My guess why speech recognition is so difficult is basically that there
MR>is a lot of information in the audio signal. There may be any number of
MR>quirks in the voice which can make it unrecognizable for the program
MR>comparing between two individuals. The program has to filter for pitch,
MR>tonal quality, speed, inflection, fricatives, and that's just a start.

MR>One can't simply match up a waveform with another but rather there is a
MR>lot of computation done on-the-fly in analyzing it to extract the
MR>information sought from a forest of waveform information. Text to speech
MR>in contrast is almost child's play since the information is already
MR>known and data is simply output from pre-existing lookup tables.

Another tid bit on voice recognition. Most of the
programs will get a considerable higher percentage of
accuracy when interpreting a female voice than when trying on
a male voice.

Jay
---
 þ OLXWin 1.00b þ Frozen diet dinners are good:but get 2-they're small
* Origin: Try Our Web Based QWK: DOCSPLACE.ORG (1:123/140)
SEEN-BY: 633/267 270
@PATH: 123/140 500 106/1 379/1 633/267

SOURCE: echomail via fidonet.ozzmosis.com

Email questions or comments to sysop@ipingthereforeiam.com
All parts of this website painstakingly hand-crafted in the U.S.A.!
IPTIA BBS/MUD/Terminal/Game Server List, © 2025 IPTIA Consulting™.