Skip to contentSkip to conversation

Speech Recogniton

DISCUSSION Michael Norton 4 participants 4 messages Sep 14, 2000 – Oct 9, 2000

lore
Share & tools
lore
#1 lore raw
Quoted 4 lines · Michael Norton

hi,

Do any libraries exist that allow me to use the Apple Speech recognition hardware under LinuxPPC?

Sorry, there is no hardware (aside from a simple digitizer to capture the sound of course). Apple's speech recognition is 100% software.

The original implementation on 68K Macs used an AT&T DSP coprocessor to do the crunching because the 68K didn't have enough oomph to do it by itself, but the PowerPC implementation relies on nothing but the CPU.

Tim Seufert

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

#2 lore raw
Quoted 7 lines · Timothy A. Seufert
Quoted 4 lines · Michael Norton

hi,

Do any libraries exist that allow me to use the Apple Speech recognition hardware under LinuxPPC?

Quoted 8 lines

Sorry, there is no hardware (aside from a simple digitizer to capture the sound of course). Apple's speech recognition is 100% software.

The original implementation on 68K Macs used an AT&T DSP coprocessor to do the crunching because the 68K didn't have enough oomph to do it by itself, but the PowerPC implementation relies on nothing but the CPU.

on the off chance that you can get sound input
to work under LinuxPPC (i have a pimiso pb so it cant AFAIK) try checking on freshmeat. There are a few programs that can do voice control. One of them is called voicecontrol i think.

I have only played with festival which can to Text to speech.

-ian reinhart geiser

.---------------------------------------------------.

| Ian Reinhart Geiser - Computer Engineering Student|
| HOME: <[email protected]>                          |
| WORK: <[email protected]>                      |
| WEB:  http://www.msoe.edu/~geiseri                |

`---------------------------------------------------'

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

#3 lore raw

Check out the Speech Recognition Package on Source Forge. "Sphinx 2" It was originally developed at Carnegie Mellon University. If you could see the code of most commercial Speaker Independant Systems, you'd likly see Sphinx running in there. I believe Apple's system was developed from sphinx. could be way off here.

http://sourceforge.net/projects/cmusphinx/

I understand it compiles cleanly on Linux PPC.

joshua

Excerpts from internet.computing.linux-ppc: 15-Sep-100 Re: Speech Recogniton by "Geiser, Ian"@msoe.edu

Quoted 6 lines
Quoted 6 lines
Quoted 4 lines · Michael Norton

hi,

Do any libraries exist that allow me to use the Apple Speech recognition hardware under LinuxPPC?

Quoted 8 lines
Quoted 8 lines

Sorry, there is no hardware (aside from a simple digitizer to capture the sound of course). Apple's speech recognition is 100% software.

The original implementation on 68K Macs used an AT&T DSP coprocessor to do the crunching because the 68K didn't have enough oomph to do it by itself, but the PowerPC implementation relies on nothing but the CPU.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/