Voice Recognition

phantom_cyph · 06-19-2007, 08:57 AM

Is there a program for voice recognition in Linux with which I can set it up to open applications and type in word documents via voice commands? (free and/or open source)

PatrickNew · 06-19-2007, 12:47 PM

Try perlbox. Its a frontend to Sphynx. I've had better sucess with the frontend than just with Sphynx.

phantom_cyph · 06-19-2007, 11:00 PM

OK. How do I configure it to use my Labtech Desk mic 524? (havent installed drivers or anything)

PatrickNew · 06-20-2007, 12:04 AM

Doesn't your microphone just plug in to the microphone socket on the audio card? It shouldn't require any drivers. I'm unsure about this, my microphone just worked out of the box.

phantom_cyph · 06-20-2007, 09:03 AM

Uh...I have the 'vocab' in I want. I didn't bother with having it talk to me, just me talking to it... but is it just supposed to work? How do I set it up to start "listening" for my keywords?

It says listening, but even talking into my mike, it doesn't do anything...kate just opened randomly and I never said editor.

phantom_cyph · 06-20-2007, 01:53 PM

Anybody know how to get this running?

frenchn00b · 11-12-2007, 12:31 PM

Hi,

Is there any progresses in voice recognition ?

I have sphinx running but it cannot recognize a s**t (or once every 50 words uttered), and uses full CPU for not much, actually. Other sphinx version s¤¤¤s rather with the java, environment, that is not linux installable with simple ./configure ... Wish it might be better with Linux...
--
help:
* can be either h or i chars. Up to you to find the right order

¤ with other chars

PatrickNew · 11-12-2007, 01:10 PM

@ frenchnoob:

This has been a point of curiosity for me. I assume from your handle that you are a native French speaker. I was under the impression that techniques required for speech-recognition would vary widely between languages and that consequently Sphinx was an "English-only" product, however I've never really tried it. Have you tried Sphinx in French or only in English? Does/can it produce results in any other language, or have you ever tried?

frenchn00b · 11-12-2007, 01:28 PM

Quote:

Originally Posted by PatrickNew

@ frenchnoob:

This has been a point of curiosity for me. I assume from your handle that you are a native French speaker. I was under the impression that techniques required for speech-recognition would vary widely between languages and that consequently Sphinx was an "English-only" product, however I've never really tried it. Have you tried Sphinx in French or only in English? Does/can it produce results in any other language, or have you ever tried?

Hi, I indeed have a friend who use dragon naturally speaking 9 for his work (he is doctor), and in french it is remarkable indeed. It gets almost 90pct. In english the results are still very okay. We just get into troubles with 'a' 'the' and simple words, that might be for you very simple to utter.

lol
sphinx french ? I might seek for that.
edit, found: http://cmusphinx.sourceforge.net/
would you know the file I should use/download ?

Which sphinx are you using ?
The only sphinx that is available in repos. is sphinx2-bin, and this one is not so easy to make it work. Are you using the usb mic with sphinx2, you ?

Code:

$ sphinx2-demo

is pretty cool to try (that it is not working for us)

frenchn00b · 11-12-2007, 01:41 PM

fr:http://doc.ubuntu-fr.org/reconnaissance_vocale
but not so great

this one is cool :
http://www.youtube.com/watch?v=w1pyw.../gnome/page/2/

frenchn00b · 11-12-2007, 01:55 PM

Ok, just installed.

The how to for Debian, etch stable, Users:

Code:

cd /tmp
wget "http://www.comp.ufscar.br/~raphael18/gnome-voice-control-0.2.tar.gz"
wget -N "http://www.comp.ufscar.br/~raphael18/gnome-voice-control-0.2.tar.gz"
tar xvpfz gnome-voice-control-0.2.tar.gz
cd gnome-voice-control-0.2
su 
apt-get -f -y install    automake-1.9
apt-get -f -y install  autotools-dev
apt-get -f -y install  debhelper
apt-get -f -y install  gnome-doc-utils
apt-get -f -y install  libatspi-dev
apt-get -f -y install  libgnomeui-dev
apt-get -f -y install  libgstreamer0.10-dev
apt-get -f -y install  libgtk-dev
apt-get -f -y install  libpanel-applet2-dev
apt-get -f -y install  libsphinx2-dev
apt-get -f -y install  libwnck-dev
apt-get -f -y install  libxml-parser-perl
apt-get -f -y install  scrollkeeper
exit
sh ./configure  --prefix=/usr
make 
sudo make install
echo "Enjoy gnome voice recognition"

Here we go !
Let's try the guy...

I made a deb for debian etch stable here:
http://yellowprotoss.ye.funpic.org/website/debs/
--

frenchn00b · 11-12-2007, 02:08 PM

how to start gnome voice recognition btw ?

it only contains stuffs in /usr/local

thank you

frenchn00b · 11-12-2007, 02:18 PM

Code:

The panel encountered a problem while loading "OAFIID:GNOME_VoiceControlApplet".

Do you want to delete the applet from your configuration?

(Damn gnome)

frenchn00b · 11-12-2007, 02:48 PM

rebooted ;

Code:

 rm -rf ./.gconf/apps/gnome-settings/gnome-panel

did it.

Okay; gnome voice control
http://raphaelnunes.wordpress.com/20...-release-gsoc/
works with gnome session:
'run terminal' works
'minimize window' works

but "next window"
and "close window" crash the gnome sessino. I have to then ctrl alt backspace

--
Since I do not use gnome session, can I use this with openbox ??

my .xinitrc

Code:

gnome-panel &
exec openbox

The idle remains, with openbox, nothing like LISTENING ...
so it means that gonme session is able to use my USB mic, but with Openbox not at all.

Which process shoudl I start so that the USB mic is seen in Openbox ??

what should I modify so that :

RUN TERMINAL : does sh ~/myscript01.sh
CLOSE WINDOW : does sh ~/myscript02.sh
...
how to recnofigure the beast ?

Sort of, I recognize pretty well

I wonder why not much more users use it .

it is so simple to install & pretty nice already

frenchn00b · 11-12-2007, 02:54 PM

I went to gnome session that has usb mic working.

Which of this process should I have in my openbox session, so that gnome voice control works ?
thanks

Code:

   PID TTY      STAT   TIME COMMAND                                                                                                  
   0:00 bash                                                                                                                         
     0:00 /usr/bin/gnome-session                                                                                                     
    0:00 /usr/bin/ssh-agent /usr/bin/dbus-launch --exit-with-session /usr/bin/gnome-session                                          
    0:00 /usr/bin/dbus-launch --exit-with-session /usr/bin/gnome-session                                                             
    0:00 /usr/bin/dbus-daemon --fork --print-pid 4 --print-address 6 --session                                                       
    0:00 /usr/lib/libgconf2-4/gconfd-2 5                                                                                             
    0:00 /usr/bin/gnome-keyring-daemon                                                                                               
    0:00 /usr/lib/bonobo-activation/bonobo-activation-server --ac-activate --ior-output-fd=18                                        
    0:00 /usr/lib/control-center/gnome-settings-daemon --oaf-activate-iid=OAFIID:GNOME_SettingsDaemon --oaf-ior-fd=26                
   0:00 /usr/bin/metacity --sm-client-id=default0                                                                                    
   0:01 gnome-panel --sm-client-id default1                                                                                          
   0:00 nautilus --no-default-window --sm-client-id default2                                                                         
    0:00 /usr/bin/python2.4 /usr/bin/gnome-osd-event-bridge --sm-disable                                                             
   0:00 /usr/lib/gnome-vfs-2.0/gnome-vfs-daemon --oaf-activate-iid=OAFIID:GNOME_VFS_Daemon_Factory --oaf-ior-fd=32                   
    0:00 /usr/lib/gnome-panel/wnck-applet --oaf-activate-iid=OAFIID:GNOME_Wncklet_Factory --oaf-ior-fd=34                            
   0:00 /usr/lib/gnome-panel/clock-applet --oaf-activate-iid=OAFIID:GNOME_ClockApplet_Factory --oaf-ior-fd=38                        
   0:00 /usr/lib/gnome-applets/gweather-applet-2 --oaf-activate-iid=OAFIID:GNOME_GWeatherApplet_Factory --oaf-ior-fd=39              
    0:00 /usr/lib/gnome-panel/notification-area-applet --oaf-activate-iid=OAFIID:GNOME_NotificationAreaApplet_Factory --oaf-ior-fd=38
   0:00 /usr/libexec/voice_control_applet --oaf-activate-iid=OAFIID:GNOME_VoiceControlApplet_Factory --oaf-ior-fd=43                 
    0:00 /usr/lib/nautilus-cd-burner/mapping-daemon                                                                                  
    0:00 /usr/lib/at-spi/at-spi-registryd --oaf-activate-iid=OAFIID:Accessibility_Registry:1.0 --oaf-ior-fd=48                       
    0:00 /usr/bin/Eterm