=================================================

=================================================

Introduction

Vox launcher is a simple speech-to-command program in Python that uses the Google's Speech Api to convert speech into text and then execute a command on the local machine.

License

This Program is distributed under the Gnu General Public License version 3. See the file COPYING for details.

Installation

Please read the INSTALL file located on the current folder.

Runtime Requirements

This program is designed to run on Linux and depends on:

• python-gtk2(>=2.16)
• python-dbus
• python-alsaaudio(>=0.5)
• python-xlib
• python-gobject-2(>=2.16)
• python-simplejson
• pynotify
• speech-dispatcher
• python-speechd
• xautomation
• python-sparqlwrapper
• sox

If you are using ubuntu you can simply run this command:

sudo apt-get install python-dbus python-alsaaudio python-simplejson python-notify python-speechd speech-dispatcher xautomation python-sparqlwrapper

Usage

Just run vox-launcher and speak! Wait a bit and the related action will be done. When the microphone in panel's notification area is red the program is not ready to listen you; please wait the green light before speak an other time.

Examples

vox-launcher can be used to launch processes:

"open firefox"

Vocal Command list

You can see the default command list handled by vox launcher reading the left column of your language configuration file (e.g. /etc/vox-launcher/vox_launcher_it.conf). If the recognized words are not declared in the configuration file vox launcher will try to launch the recognized text as is.

Query/Answering

If the recognized test is not a command the program interpret your request as a query and it try to answer using wikipedia data.

Now you can make query with these formats:

• Who is ? (e.g. Who is Barack Obama?)
• What is ? (e.g. What is Freemasonry?)

In italian language

• Chi è (es. Chi è Beppe Grillo?)
• Cosa è (es. Cosa è Massoneria?)

also these shortly form are allowed:

• (e.g. Barack Obama)
• (e.g. Freemasonry)

Fast web access

You can open website whith this command:

• Go to "website" (e.g. Go to wikipedia)

In italian language:

• Vai su (es. Vai su wikipedia)

Text Insertion

You can insert text with this vocal command:

• Write (e.g. Write George Washinghton)

In italian language:

• Scrivi (e.g. Scrivi Giuseppe Garibaldi)

Vocal pointer

With vox-launcher you will able to move the pointer with voice, the mouse is not required anymore.

Talk the keyword "grid" (or "griglia" if the italian language has been selected) a 3x3 grid on your screen will appear:

1 2 3 4 5 6 7 8 9

Please speak the cell number where do you want move the cursor. You can continue to reiterate the procedure in the selected cell.

When you will are upon the desired screen point you can click the pointer using these vocal command:

• "click"
• "double click" (or "doppio click" if the italian language has been selected)
• "right click" (or "click destro" if the italian language has been selected)

Customization

Vox launcher is highly configurable modifying some configuration file located under the /etc/vox-launcher folder. In order to customize the vox laucher we must edit the the correspondent language file. For example to customize the recognition in italian language we can modify the /etc/vox-launcher/vox_launcher-it_IT.conf file.

The semantic of the conf file is the simplest as possible following the behavioural stimulus-response pattern:

[detected word] -> [command to run]

e.g. browser -> chromium-browser

Voice launcher when will detect the browser world will launch the chromium-browser command. You can change the default browser from chromium to firefox simply changing the previous line into:

browser -> firefox

Info

• To get info about the tool, please contact:

  pilolli.pietro@gmail.com

• To report bugs, please contact:

  pilolli.pietro@gmail.com

Have fun!

=== EOF