About the essential patent of MELP
10105...@163.com
Recently, I'm working on the promotion of a kind of digital
walkie-talkie system standard. And I need to find the patent lists of
several vocoders which can work at or below 4kbps. Among the
candidates I think MELP is good choice. Although I know TI is the
patent holder, the amount of patent of TI is quite huge. I really
can't assure I can find the full patents.
So, is there anyone can tell me the patent list of MELP,
thanks very much!
10105...@163.com
I only find one patent "Mixed excitation linear prediction with
fractional pitch", I don't know if that is all?
Gonzo
MELP / MELPe is an excellent choice for state of the art vocoder at
rates below 4kbps.
Indeed, TI holds several patents in MELP including the above one (switch
predictive VQ etc.). They typically waive their royalties once their
DSP is used.
Also, if you intend to use the much better-quality vocoder:
enhanced-MELP (a.k.a. MELPe), be aware that additional companies hold
IPR in different elements in it, including are Microsoft, Compandent,
AT&T, and Thales Group. You may contact each IPR holder to find out
their policy. Out of the above IPR holders, the only company that
actively supports and develops MELPe related OEM products is Compandent.
They may also help you in contacting the other IPR holders, see their page:
http://www.MELPe.com
http://www.compandent.com/products_melpe.htm
http://www.Compandent.com/melpe_faq.htm
http://www.compandent.com/products_flexi232.htm
10105...@163.com
> 1010520...@163.com wrote:
> > On 3月6日, 上午11时52分, 1010520...@163.com wrote:
> >> Hello:
> >> Recently, I'm working on the promotion of a kind of digital
> >> walkie-talkie system standard. And I need to find the patent lists of
> >> several vocoders which can work at or below 4kbps. Among the
> >> candidates I think MELP is good choice. Although I know TI is the
> >> patent holder, the amount of patent of TI is quite huge. I really
> >> can't assure I can find the full patents.
>
> >> So, is there anyone can tell me the patent list of MELP,
> >> thanks very much!
>
> > I only find one patent "Mixed excitation linear prediction with
> > fractional pitch", I don't know if that is all?
>
> rates below 4kbps.
>
> Indeed, TI holds several patents in MELP including the above one (switch
> predictive VQ etc.). They typically waive their royalties once their
> DSP is used.
>
> Also, if you intend to use the much better-quality vocoder:
> IPR in different elements in it, including are Microsoft, Compandent,
> AT&T, and Thales Group. You may contact each IPR holder to find out
> their policy. Out of the above IPR holders, the only company that
> They may also help you in contacting the other IPR holders, see their page:http://www.MELPe.comhttp://www.compandent.com/products_melpe.htmhttp://www.Compandent.com/melpe_faq.htmhttp://www.compandent.com/products_flexi232.htm- 隐藏被引用文字 -
>
> - 显示引用的文字 -
Thanks for your reply. I've also found the other patents US6463406 for
the MELP.
fizteh89
>
> > - 显示引用的文字 -
>
> Thanks for your reply. I've also found the other patents US6463406 for
>
> - Show quoted text -
US6463406 is just a pitch period determination using cross-correlation
- it has been made obsolete by US7124075
Compandent doesn't have any patents on MELP/MELPe, meaning that as
long as you can write your own code you are in the clear... (Sorry
Gonzo a.k.a Oded :-)
It's really just US5699477 you are up against, you know... those big
lawyers from Texas in cowboy boots... Don't mess with those guys -
they managed to extract patent royalties even in Japan - the
unthinkable feat...
Regis
>
> > > - 显示引用的文字 -
>
> > Thanks for your reply. I've also found the other patents US6463406 for
> > the MELP.- Hide quoted text -
>
> > - Show quoted text -
>
> US6463406 is just a pitch period determination using cross-correlation
> - it has been made obsolete by US7124075
> Compandent doesn't have any patents on MELP/MELPe, meaning that as
> long as you can write your own code you are in the clear... (Sorry
> Gonzo a.k.a Oded :-)
> It's really just US5699477 you are up against, you know... those big
> lawyers from Texas in cowboy boots... Don't mess with those guys -
> they managed to extract patent royalties even in Japan - the
>
> - Show quoted text -
It seems that the one asking the question found me at the EPO, that
was my answer:
A mixed excitation linear prediction (MELP) coefficient vocoder is
described in an article by A. McCree, et al. entitled "A Mixed
Excitation LPC Vocoder Model for Low Bit Rate Speech Coding", in IEEE
Trans. on Speech and Audio Proc., Vol. 3, No. 4, July 1995. This is
also described in the PhD Thesis of A. V. McCree for Georgia Institute
of Technology, August 1992.
The first TI patents building up on McCree's work seem to be US5699477
and US5794180
;-)
LittleStar
answers such as the one below won't be much helpful for defense in
court... :)
fizteh89
as I have no understanding of the subject, and I won't be helpful in
case you'd be sued for infringement by any of the 5 IP holders. You'd
better refer to:
http://www.MELPe.com/
http://www.compandent.com/products_melpe.htm
MELPe Frequently Asked Questions (FAQ):
http://www.compandent.com/melpe_faq.htm
Dmitry Terez (te...@soundmathtech.com)
+1.856904xxxx
Fax: +1.856581xxxx
P. O. Box 846
Cherry Hill, NJ 08003
US