Sign in

username:

password:



Not a member?

Search speechcoding



Search tips

Subscribe to speechcoding



speechcoding by Keywords

ACELP | ADPCM | AMBE | AMR | AMR-NB | CELP | Codebook | DTMF | G.723 | G.726 | G.729 | GSM | Interpolation | LPC | LSF | LSP | MELP | PCM | Perceptual | Pitch | PSOL | QCELP | Quantization | SMV | VAD | Vocoder

Sponsor

NEW! TMS320C6474: 3x the performance. 1/3 the cost. Three 1 GHz cores on 1 chip.

Discussion Groups

Discussion Groups | Speech Coding | Re: Queries regarding G.723.1

Technical discussions related to Speech Coding (all itu and other vocoders, ACELP, CELP, AMR, etc)

  

Post a new Thread

Queries regarding G.723.1 - Nihal Ahmad - Mar 28 6:12:00 2001



Hi friends I am working on G.723.1 Annex B in
simulink/Matlab environment. I have some doubts which
I am giving below.
If any body of you know the good material regarding to
G.723.1 Annex B.
Let me know.
Regards
Nihal

----------------------------------------------

The standard is specifying some equations in the
MP-MLQ block ,but it is not clear what different
operations are being performed by the block.

1)I would like to know what are all the inputs to the
MP-MLQ/ACELP block other than h(n) and r(n).
& what are the transmitted parameters

2) what are the distinct operations that are performed
by this block(are we finding the excitation vector
v(n) or it is accessec by this block during
computation).

3)the estimated gain is Gmax is found by equation (25)
we are quantizing it using logarithmic quantizer of 24
steps of 3.2db each.but it says around this Gmax(q) we
find more values in the range –3.2db to +6.4db. how
many values are we finding and how?

4)the last para of the block says if the value of L0
or L2 is less than 58 we use a train of dirac
functions instead of a single dirac function.please
explain the reason and theory behind this

Please Provide hierarchical flow diagram/chart/steps
for the MP-MLQ & ACELP blocks .ANSI C code provides
functions that do different tasks but they are so much
interconnected that it is very difficult to understand
the code without going thru the whole C code.What I
want is logical flow in the functions and briefly
describing the those functions. __________________________________________________

Attachment (not stored)
queries.doc
Type: application/msword




(You need to be a member of speechcoding -- send a blank email to speechcoding-subscribe@yahoogroups.com )

Re: Queries regarding G.723.1 - Author Unknown - Sep 30 18:31:00 2001


Hi all,

I am working in a project on g723.1.I wanted to know the purpose of
converting the lpc coefficients to lsp frequencies.Also in the 4
subframes that we are creating why do we convert the 10 lpc
coefficients coming out of the last subframe only to 10 lsp
frequencies.Again we are converting these lsp frequencies back to the
lpc coefficients . while transmitting are we transmitting all the 40
lpc coefficients or just the lsp frequencies.please help me to
understand the process clearly.

Thanks and regards,
Karthikeya raghav
--- In speechcoding@y..., Nihal Ahmad <q_nihal@y...> wrote:
> Hi friends I am working on G.723.1 Annex B in
> simulink/Matlab environment. I have some doubts which
> I am giving below.
> If any body of you know the good material regarding to
> G.723.1 Annex B.
> Let me know.
> Regards
> Nihal
>
> ----------------------------------------------
>
> The standard is specifying some equations in the
> MP-MLQ block ,but it is not clear what different
> operations are being performed by the block.
>
> 1)I would like to know what are all the inputs to the
> MP-MLQ/ACELP block other than h(n) and r(n).
> & what are the transmitted parameters
>
> 2) what are the distinct operations that are performed
> by this block(are we finding the excitation vector
> v(n) or it is accessec by this block during
> computation).
>
> 3)the estimated gain is Gmax is found by equation (25)
> we are quantizing it using logarithmic quantizer of 24
> steps of 3.2db each.but it says around this Gmax(q) we
> find more values in the range –3.2db to +6.4db. how
> many values are we finding and how?
>
> 4)the last para of the block says if the value of L0
> or L2 is less than 58 we use a train of dirac
> functions instead of a single dirac function.please
> explain the reason and theory behind this
>
> Please Provide hierarchical flow diagram/chart/steps
> for the MP-MLQ & ACELP blocks .ANSI C code provides
> functions that do different tasks but they are so much
> interconnected that it is very difficult to understand
> the code without going thru the whole C code.What I
> want is logical flow in the functions and briefly
> describing the those functions. > __________________________________________________




(You need to be a member of speechcoding -- send a blank email to speechcoding-subscribe@yahoogroups.com )