[sndtools] SPEECH PROCESSIGN from INDIA

Ananya Misra amisra at CS.Princeton.EDU
Fri Sep 17 18:37:31 EDT 2010


Hi Raktim,

The canonical paper on LPC is probably "Speech Analysis and Synthesis
by Linear Prediction of the Speech Wave" by B. Atal, Journal of the
Acoustical Society of America (1970/71). Perhaps others have
suggestions for more recent tutorials and papers.

The relevant visualization software in the sndtools package is rt_lpc:
http://soundlab.cs.princeton.edu/software/rt_lpc/

It's open source, so you can also enjoy the source code in this download:
http://soundlab.cs.princeton.edu/software/rt_lpc/files/rt_lpc-1.0.tgz

Thanks!
-Ananya

On Fri, Sep 17, 2010 at 2:49 AM, Raktim Haldar
<raktim.halder21 at gmail.com> wrote:
> To
> Princeton University,
>  I am an Electronics Engineer by profession. I am from India , and
> wish to do some works on the very interesting topic,' REAL-TIME TEXT
> INDEPENDENT SPEECH RECOGNITION'. This is really challenging as no
> system can recognize speech with more than 80% accuracy. We
> individually have done some works on this topic using MATLAB. I found
> some usefulness of LINEAR PREDICTIVE CODES in speech synthesis and
> voice recognition, But due to lack of facilities and resources we can
> not progress especially on this particular aspect.
>               I found that you are one of the pioneers of this
> field. So I like to ask your help. Can you provide me some very useful
> papers and tutorials on this 'LPC' ? It will be also very helpful if
> you provide me a .EXE file or software which can show me a demo of
> 'LINEAR PREDICTIVE CODE ANALYSIS' and its 'RESIDUE' analysis.
>               Hope you can find a little bit time from your busy
> schedule for me. I need your help badly.
> Thank You very Much.
> Yours faithfully
> Raktim haldar
> India.
> _______________________________________________
> sndtools mailing list
> sndtools at lists.cs.princeton.edu
> https://lists.cs.princeton.edu/mailman/listinfo/sndtools
>


More information about the sndtools mailing list