Ubuntu QA:
BlogBrainstormPackage status
Log in
Ubuntu QA
The Ubuntu community has contributed 21545 ideas, 132412 comments, 2606687 votes
Idea sandbox Idea sandbox
Popular ideas Popular ideas
Ideas in development Ideas in development
Implemented ideas Implemented ideas
Idea #1826: Increase accessibility with voice recognition

bug This idea was marked as already implemented the 9 July 11. Available starting Ubuntu 7.10 Gutsy Gibbon.
Written by TeyeDoubleGuhRrr the 29 Feb 08 at 21:08. Category: Others. Related project: Nothing/Others. Status: Already implemented
Rationale
Having the ability to control the Ubuntu / Kubuntu interface via voice commands and the ability to dictate into applications would contribute immensely to the accessibility of the product.

Currently the Sphinx project of CMU (http://cmusphinx.sourceforge.net/html/cmusphinx.php) is one of the better known open-source speech recognition projects. Perhaps working with them to enhance their functionality would be beneficial to the Ubuntu meme as a whole.
Tags: (none)

540
votes
closed
Solution #1: Auto-generated solution of idea #1826
Written by TeyeDoubleGuhRrr the 29 Feb 08 at 21:08.
Ubuntu Brainstorm was updated in January 2009. Since the idea #1826 was submitted before this update, its rationale and solution are not separated. Please vote accordingly, and if you have the necessary rights, please separate the rationale from the solution. Thanks!
11
votes
closed
Solution #2: Start with just commands.
Written by teeks99 the 6 Apr 09 at 16:14.
I love the idea...I just watched Iron Man last weekend, and now I really want a "Jarvis" that will do what I tell it to do!

That said, getting the OS to support doing *everything* with voice commands doesn't sound like something that will be easy to accomplish anytime in the near future. I think this would be even better if it could be scoped down to a list of commands that it understands...like "open firefox. Open bookmark google mail", or "open amarok. start playlist recent songs".

Having just this simple kind of command support would open up all sorts of new uses for ubuntu. Imagine using your smart phone without having to type into a tiny keyboard, or navigate across a diminutive screen to find the program you want. Or imagine calling up your home PC over ekiga and saying "mythtv record show the office at 8pm"!

The possibilities are endless!
5
votes
closed
Solution #3: Put a GUI for speech recognition out there
Written by ohadle the 26 Mar 10 at 09:44.
As I understand, one of the main problems of open source speech recognition is not having enough voice samples.

Getting a basic front-end for voice recognition, that would make contributing to databases such as Voxforge easy, could push the field forward.

I propose a very basic GUI at first, enabling say speech-to-text and basic macros ("Open Firefox" etc.). This GUI should enable easily typing in the voice command you just said and submitting to VoxForge.
3
votes
closed
Solution #4: put SPHINX installed in distros
Written by mitcoes the 25 Oct 10 at 15:01.
Even it is at USC, being installed as default in Distros, UNE and Desktop, will be the future for tablets, and touchscreens desktops future "normal" users.

And of course this program will improve a lot if the bese of users is incremented.

Propose your solution

Attachments
No attachments.


Duplicates


Comments
robin wrote on the 29 Feb 08 at 21:25
I voted in favor, even though gnome-voice-control already exists and is available in the ubuntu repositories. However everyone who is interested in this functionality should go to http://www.voxforge.org/ and submit speech.

Without enough (GPL) speech speech recognition on linux will simply not happen even though there are many people out there who'd like to use speech recognition and there are programs out there too.

We need training data!!! I submitted hours myself, but we need many voices to make it 'speaker independent' and lots of speech per speaker helps as well. This is not something ubuntu can do for you without your speech.

ethana2 wrote on the 2 Mar 08 at 09:01
Finally, someone else sees it. *votes up*
You just saved me five minutes.

Not only do we need them, they need us. Get Ubuntu users involved. I think most of us are capable of talking. The more ways to contribute, the better!

Get festival in too.

jogo wrote on the 10 Mar 08 at 15:38
I dont linke gvc style, I would like more a voice control just based on the user like cvoicecontrol. And then you dont need speech libraries.

pierre.slamich@gmail.com wrote on the 1 Jul 08 at 23:40
-Packaging Quickstart from http://www.voxforge.org/ would greatly help getting an open corpus.

-Simon Listens is a promissing effort: http://simon-listens.org/

-GnomeVoiceControl seems to be making progress , but on a slowish basis

andreas_k wrote on the 13 Sep 08 at 02:53
It would be really great, when you can use (your own voice) for starting programms like
firefox or internet (and firefox will be starten)
amarok next (and in amarok the next song will be played)
or a nice feature in impress, okular, evience, ...
a presentation will be starten and when i say: next slide the next slide or side will be come.

I`d like to use my voice like shortcards. at the start the programm ask you how to start the programms and the programm also ask you some often used words (like next, ...) and in the programs you can put alternative words for next. like you can make it with shortcards in kde applications.

I`think it will be realy nice, when you have an presentation and when you say next slide the computer "works" for you.

sinbad wrote on the 19 Feb 09 at 03:50
If it can be done in vista why not ubuntu!

sheep1364 wrote on the 20 May 09 at 04:12
What happens when I want to issue an administrative command that requires the root password via voice control? I don't want to say my password out loud for everyone to hear, is there a way to implement security features into voice recognition so that if someone does hear me say my password or passphrase they won't be able to use it to gain administrative access?

One idea would be to bind a separate voice command string to the root password, i.e when I say "Alpha 8 Gecko" the system will recognize it as the voice version of the root password. So basically you would have 2 different root passwords, one you can type in and one you say.

OpenNingia wrote on the 20 May 09 at 10:16
why not neural interface then?

:P

cheesehead (Brainstorm admin) wrote on the 9 Jul 11 at 16:07
Several voice-recognition packages are already in the Ubuntu repositories.
Improvement is ongoing. Your contributions are welcome.


Post your comment