![]()
|
I would like my computer to have the ability to read to me. Better yet, have the capacity to understand commands and keywords, issued verbally and then take that information, go out on the internet, research and compile related data into a dissertation which is comprehensive and intelligible.This exercise had little to do with that. This project was merely a study in recording text to speech on the fly as well as a diversion into just diddling around. When I don't feel like doing anything, but can't find myself not doing anything, I do stuff like this. The Macintosh has had the ability to read text documents since its beginnings.
What was of interest here was being able to easily record the text to
speech and creating an AIFF file which would lend itself to further processing.
WireTap by
Ambrosia Software is the
first program I have found that would record any
audio being played on the Mac and does so with a single click. By default
it opens the recorded files in QuickTime. I changed this to Bias-Inc's
Peak 3.2 for a better visual and greater editing and processing capabilities.
System Preferences --> Speech was left open so I could quickly change
the default voice. WireTap is the little gui at the top of the screen
toward the center. Record, stop and pause buttons, displays elapsed time
and file size. The preferences is open to show settings; even though you
can't read them. I was going to link a larger screen shot but it's easy
enough to figure out. The entire Scene 1 text reading was about 2'50" with 38 lines. There was very little processing in Peak, the duration of a couple of lines were increased, gain was increased in a couple of places to accent a syllable and there were a couple of modest frequency adjustments on words. The AIFF was compressed to mp3 using iTunes. Scene 1 Monty Python and the Holy Grail (click da button) |
|
| The Movie, Scene 1: King Arthur: Graham Chapman Patsy: Terry Gilliam Soldier #1: Michael Palin Soldier #2: John Cleese |
The Mac, Scene 1: |
|
|
A Brief History of Time, Documentary. 1992 This program was on the Science Channel today. Hawking's theories are a backdrop in this documentary which is primarily a retrospect of his life. All of the dialogue produced by Stephen's computer is read by Fred. The computer is an IBM PC, probably a 80486 processor. The Pentium processor wasn't introduced until 1993. The sound of Fred reading Hawking's text may have been dubbed on the video and it is doubtful that there is much real time in the readings. The first Sound Blaster card was introduced in 1992. Anyway, it was about the same quality then as it is today. His software was a text editor and dictionary with text to speech capability, that's about it. There is a simple hand held mouse device he uses to navigate. What we need here is a way to tap the verbal brain activity and convert it digitally and then process with a speech synthesizer. Why is this still so far away? Doesn't really matter much in this case. Mr. Hawking is in his 50's. Great thinkers peak out in their early 30's while great artists peak out after they're dead. |
||