Speex dev - Aug 2004 - speex_encode / speex_encode_int ... already sorry =)

Hi, i know this question has already been asked, i've checked mailing 
list logs but ...
I didn't figure out the solution for my question.

I want to use basic functionnality of speex, only VAD, no pre/post 
filter, even no VBR. But i *really* want to use less mem/cpu i can.
So in which format should i pass data to speex_encode(_int) , float or 
short ?
I quickly looked at the source code and i saw the speex_encode doing 
float to short convertion but it has protected by some #ifdef so ..

Easy question, i want to:
    - pass buffer to Speex in its native format
    - encode fast with low resource, avoid buffer convertion ...

peex_encode_int / speex_encode ?

thank you :)
-- cl
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to
'speex-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

> speex_encode_int / speex_encode ?
Please use speex_encode_int unless you have a good reason not to.

        Jean-Marc


-- 
Jean-Marc Valin
http://www.xiph.org/~jm/
LABORIUS
Université de Sherbrooke, Québec, Canada


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 190 bytes
Desc: Ceci est une partie de message numériquement signée.
Url :
http://lists.xiph.org/pipermail/speex-dev/attachments/20040425/ba6d850c/signature-0001.pgp

On Sun, Apr 25, 2004 at 04:18:16PM +0200, crazylord
wrote:
> Hi, i know this question has already been asked, i've checked mailing 
> list logs but ...
> I didn't figure out the solution for my question.
> 
> I want to use basic functionnality of speex, only VAD, no pre/post 
> filter, even no VBR. But i *really* want to use less mem/cpu i can.
> So in which format should i pass data to speex_encode(_int) , float or 
> short ?
> I quickly looked at the source code and i saw the speex_encode doing 
> float to short convertion but it has protected by some #ifdef so ..
> 
> Easy question, i want to:
>    - pass buffer to Speex in its native format
>    - encode fast with low resource, avoid buffer convertion ...
> 
> speex_encode_int / speex_encode ?
peex_encode_int()

but it will only work with version 1.1 (and maybe future versions), if
you want compatability with the stable version 1.0 you need to use
speex_encode().

Conrad.
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to
'speex-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

I always think short is better over float, because most of the time we don't
get float form data, from, e.g microphone / wav files

<p>-----Original Message-----
From: owner-speex-dev@xiph.org [mailto:owner-speex-dev@xiph.org] On Behalf
Of Conrad Parker
Sent: Monday, April 26, 2004 12:58 AM
To: speex-dev@xiph.org
Subject: Re: [speex-dev] speex_encode / speex_encode_int ... already sorry
=)

<p><p>--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to
'speex-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.

I have been trying to develop a dll for speex-1.1.4. But I am having problem 
putting in the VAD feature. It seems that speex has put a very low threshold 
for VAD activation. Even a blank audio file in windows contains int values 
almost all greater than 10.

And VAD comes on only when the input buffer has values less than 3 for 
wideband. Therefore any recorded file/input from mike never activates VAD in 
windows. what can I do to overcome this difficulty?

One possible solution seems like dividing all the values in input buffer by a 
suitable integer (this comes out to be atleast 190, found by trial n error). 
This lowers even the required speech too much in volume, hence not a very 
useful option. 

Other alternative seems to set a threshold for division as well. If the input 
buffer value is smaller than some threshold then divide else leave it as it 
is. This again seems to be a trial n error method for deciding the threshold. 

Hence, what other setting can be changed? Everything works fine in Linux but 
windows gives problem. what can be done?

-Abhishek 
--- >8 ----
List archives:  http://www.xiph.org/archives/
Ogg project homepage: http://www.xiph.org/ogg/
To unsubscribe from this list, send a message to
'speex-dev-request@xiph.org'
containing only the word 'unsubscribe' in the body.  No subject is
needed.
Unsubscribe messages sent to the list will be ignored/filtered.