[PD] voice regognition (speech2text)

Nicholas Ward nicholas.ward at cs.tcd.ie
Wed May 24 23:30:05 CEST 2006


i used sonic from university of colorado a few years ago. it was pretty  
quick to get running as i remember with the supplied examples. Its a  
client/server setup. There is also a supplied example client written in  
tcl/tk which was easy to hack so it sent the words the server  
recognised on to a socket and then into PD.
Of course i cant find any of this now but heres where you can get sonic  
from (you just need to register to download). I ran the server on OS X  
fine (needs a fast computer) and i also saw a friend run it very  
successfully on windows.
Hope this helps.


http://cslr.colorado.edu/beginweb/speech_recognition/sonic_main.html


On 24 May 2006, at 20:20, Erich Berger wrote:

> i used the ibm libraries (cant remember the name eather :-) years ago
> under linux quite successfully for a voice controlled robot.
> back then one had to train the engine about 20 minutes with a given
> text one had to read -
> but then it worked really well - for the person who trained it.
>
> erich
>
>
>
> ---------------------
> http://randomseed.org
>
> On Wed, 24 May 2006, Pall Thayer wrote:
>
>> You could try doing something with CMU's Sphinx. It compiled fine on  
>> OS X with no errors but I can't get it to run. Maybe someone else  
>> can:
>>
>> http://cmusphinx.sourceforge.net/html/cmusphinx.php
>>
>> Pall
>>
>> On 24.5.2006, at 12:44, Hans-Christoph Steiner wrote:
>>
>>> I would love something like that too, but I haven't had the chance  
>>> to dig for it.  I know that IBM has done a lot of research into  
>>> speech recognition, and is big into free software.  I vaguely  
>>> remember them releasing some of that stuff under a free license.   
>>> Its definitely worth some time to try digging around IBM's site.
>>> .hc
>>> On May 24, 2006, at 5:37 PM, B. Bogart wrote:
>>>> Hey all,
>>>> Does anyone have a patch that tries to convert a live audio input  
>>>> into a
>>>> text (english) that approximates the input? It need not be  
>>>> perfect.. If
>>>> anyone has a patch that picks out the vowels and prints aeiou that  
>>>> is a
>>>> good start...
>>>> Or does anyone know of a linux/osx GPL library that does this?
>>>> Thanks all!!
>>>> .b.
>>>> _______________________________________________
>>>> PD-list at iem.at mailing list
>>>> UNSUBSCRIBE and account-management -> http://lists.puredata.info/  
>>>> listinfo/pd-list
>>> _____________________________________________________________________ 
>>> _ ______
>>> "Terrorism is not an enemy.  It cannot be defeated.  It's a tactic.   
>>> It's about as sensible to say we declare war on night attacks and  
>>> expect we're going to win that war.  We're not going to win the war  
>>> on terrorism."
>>>                                     - retired U.S. Army general,  
>>> William Odom
>>> _______________________________________________
>>> PD-list at iem.at mailing list
>>> UNSUBSCRIBE and account-management -> http://lists.puredata.info/  
>>> listinfo/pd-list
>>
>>
>>
>> --
>> Pall Thayer
>> palli at pallit.lhi.is
>> http://www.this.is/pallit
>>
>>
>>
>>
>> _______________________________________________
>> PD-list at iem.at mailing list
>> UNSUBSCRIBE and account-management ->  
>> http://lists.puredata.info/listinfo/pd-list
>>
>
> _______________________________________________
> PD-list at iem.at mailing list
> UNSUBSCRIBE and account-management ->  
> http://lists.puredata.info/listinfo/pd-list
>





More information about the Pd-list mailing list