Installing Ubuntu Jaunty

This page provides instructions for downloading and installing Ubuntu Jaunty. If you read anything, read this: INSTALL JAUNTY FROM THE LIVE CD USING THE SCREEN READER OPTION. If you don't do this, audio will be messed up on your machine and speech will be choppy, unintelligible, and will cut off mid-sentence. Note also that all Ubuntu accessibility questions should be directed to the Ubuntu Accessibility List. The Ubuntu team needs to hear your comments.

Install using the Screen Reader Option

Perform these steps to boot with the screen reader. You will not get any audio notification during this process:

1. Boot the live CD - other than the CD stopping spinning, you will not get any audible notification that the system is ready.
2. The initial boot screen displays a language selection dialog. Press the escape key to dismiss the language selection dialog.
3. Press F5 to bring up the accessibility menu.
4. Press the number 3 or the down arrow key 3 times to select the screen reader option.
5. Press return to dismiss the accessibility menu.
6. Press return again to boot the machine.

When the machine boots to the desktop, Orca will automatically launch and its preferences dialog will be open. Press the escape key to dismiss the dialog. Then press Ctrl+Alt+d to select the desktop. Press down arrow once to select the installer. Press enter to launch the installer.

The installer could use some accessibility work, but by arrowing and tabbing around, you can make your way through an install.

Once the system is installed, it will play an audio prompt when the login screen is ready. Login by typing your username, return, your password, and then return once more. Orca will automatically launch.

D'Oh - I installed without the Screen Reader Option and Speech is not Working!!!

If you have not installed your machine using the blindness profile, you can perform a workaround that was posted to the Ubuntu Accessibility List:

https://lists.ubuntu.com/archives/ubuntu-accessibility/2009-May/003493.html

If this fails, simply go into the terminal (usually alt-f2 and type "gnome-terminal") and give the command "sudo apt-get remove pulseaudio", then restart gnome with the command: "sudo /etc/init.d/gdm restart". You must enter your password when you use the sudo command, usually only the first time.

Use orca with pulseaudio via speech-dispatcher

I am finding Ubuntu 9.04 x64 more stable if installed without removing pulseaudio. Unfortunately, you'll need help from a sighted user, as Ubuntu 9.04 x64 has broken speech output until you do several steps. First, I recommend following the Vinux instructions for making your Linux system more accessible in general. If you do so, you will be using the standard Vinux keybindings, which are far better than the default keybindings in any other linux distribution, IMO. The most recent Vinux roll-your-own instructions can be found here.

The key steps from the blog are:

16. Configure PulseAudio: Type 'sudo adduser vinux pulse-rt' to add yourself to the pulse-rt group to get real time access to the PulseAudio sound server. Replace 'vinix' with your user name. This step is not required, but should improve speech performance.

22. Install and configure speech-dispatcher: Run 'sudo apt-get install speech-dispatcher python-speechd' in a terminal. Once installed run 'spd-conf' in the same terminal.Accept all of the default options except type 'pulse' for the soundserver and '6560' for the synthesizer port. Speech-dispatcher should then speak if everything is going to plan! Use the 'speech-dispatcher -d' command to start up speech-dispatcher as a daemon'Open the Orca preferences window and select speech-dispatcher as the speech synthesizer and click apply.Orca should now be a lot more snappy and responsive.Finally open the Session Manager from the preferences menu, createa new entry called speech-dispatcher and type 'speech-dispatcher -d' as the command, then reboot and enjoy you new snappier speech!

Note: you must also specify 'pulse' rather than the default of 'alsa' when spd-conf asks for the default speech driver.

You then need to configure orca to output to speech-dispatcher. In orca preferences, go to "speech", then change the "Speech System" to "Speech Dispatcher", then select "OK" to save. Reboot should not be necessary.

In Ubuntu 9.04, there is no need for installing xbindkeys, as you can now add custom keybindings using System/Preferences/Keyboard Shortcuts. You must install a shortcut to restart Orca, or you will become very frustrated. Orca becomes unresponsive every few minutes, mostly because speech-dispatcher crashes. The following keybinding will restart both speech-dispatcher and Orca:

Name: Orca

Command: sh -c 'killall speech-dispatcher; speech-dispatcher -d; orca'

I also highly recommend using the other keybindings recommended in the blog, and Stormdragon's amazing time/date/weather Orca shortcuts (step 21 in the blog). I have also installed the Voxin voxin-enu-0.23 voice, which is much better in my opinion than the others. You can buy it for under $10here.

Bleeding Edge 64-bit Ubuntu Jaunty with IBM ViaVoice

First, get the latest version (version 0.23 as of this edit) of Voxin, as described above. Earlier versions wont work on Ubuntu 9.04 x64. The espeak audio driver is still not working well with gnome speech services and pulseaudio, but the voxin drivers do!

The most recent upgrades to various packages seem to make Ubuntu Jaunty x64 more stable with Orca, including compiz, which has a very nice zoom capability. First, follow these Download/Install instructions and install the latest atk, at-spi, and orca. Then, add these lines to your /etc/apt/sources-list file:

# Proposed deb http://archive.ubuntu.com/ubuntu/ jaunty-proposed main restricted universe multiverse

deb-src http://gb.archive.ubuntu.com/ubuntu/ jaunty-proposed main restricted universe multiverse

# Compiz deb http://ppa.launchpad.net/compiz/ubuntu jaunty main

deb-src http://ppa.launchpad.net/compiz/ubuntu jaunty main

Then, do the standard apt-get update, and apt-get upgrade. However, to activate the latest compiz, I also had to do 'apt-get upgrade compiz'.

At this point, I was able to use the voxin voice with Orca, pulseaudio, and Gnome Speech Services, with good performance. To get the alt-tab window switcher working reasonably well, I had to turn off fading windows in System/Preferences/CompizConfig Settings Manager. Compiz has a new bug which causes it to only allow two destops in the desktop switcher, but I can live with that for now. There is also some strange bug in the window switcher that causes the first window title to be cut short. However, overall, it's working fairly well. Moving back to Gnome Speech Services fixed two problems: speech-dispatcher was crashing every few minutes, and it seemed to ignore speech.stop() Orca commands, so you had to listen to queued text you didn't want to sometimes.