IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
documentation for vorbis encoder
kumarsvinod20
post Jan 25 2013, 10:34
Post #1





Group: Members
Posts: 5
Joined: 25-January 13
Member No.: 106176



Hi all,
I am new to this forum and currently working on developing vorbis encoder, downloaded the libvorbis-aotuv_b6.03 as i have many doubts. i would like to know where i can get related documents for vorbis compression algorithm.

Thanks in advance,
Regards
Vinod
Go to the top of the page
+Quote Post
ktf
post Jan 25 2013, 11:10
Post #2





Group: Members
Posts: 369
Joined: 22-March 09
From: The Netherlands
Member No.: 68263



Have you tried this one?

http://xiph.org/vorbis/doc/


--------------------
Music: sounds arranged such that they construct feelings.
Go to the top of the page
+Quote Post
kumarsvinod20
post Jan 25 2013, 13:10
Post #3





Group: Members
Posts: 5
Joined: 25-January 13
Member No.: 106176



QUOTE (ktf @ Jan 25 2013, 15:40) *
Have you tried this one?

http://xiph.org/vorbis/doc/


hi thanks for your reply,
yes i tried that, but i found docs related only to specifications and the compression algorithm is not explined much. Basically i am searching for some documenation which explains the whole compression algorithm (used in refernce encoder) in detail for eg: psychoacoustic model, channeel coupling and encoding of residue etc...

Thanks in advance
Vinod
Go to the top of the page
+Quote Post
Dynamic
post Jan 28 2013, 03:45
Post #4





Group: Members
Posts: 817
Joined: 17-September 06
Member No.: 35307



Why not read the Wikipedia Vorbis page's Technical section and follow the references to technical web pages and scholarly papers describing things further. These give a lot more detail. There's also some information in Xiph's CELT and Opus blogs regarding certain techniques (e.g. Constrained Band Energy, Band Folding, Pyramidal Vector Quantization etc.).

Actually, most of that's in the Specification. If you want to know about Psychoacoustic Models, the general principles are explained in numerous places with reference to other codecs, and most of these techniques are applicable to the tools available in Vorbis. You might also want to look into AoTuV source code versus libvorbis to spot differences in tuning. I think the libvorbis source code is probably pretty good documentation, then search Wikipedia, Google Scholar, and finally Google generally, to understand terms you don't recognise.


This post has been edited by Dynamic: Jan 28 2013, 04:08
Go to the top of the page
+Quote Post
kumarsvinod20
post Jan 28 2013, 07:28
Post #5





Group: Members
Posts: 5
Joined: 25-January 13
Member No.: 106176



QUOTE (Dynamic @ Jan 28 2013, 08:15) *
Why not read the Wikipedia Vorbis page's Technical section and follow the references to technical web pages and scholarly papers describing things further. These give a lot more detail. There's also some information in Xiph's CELT and Opus blogs regarding certain techniques (e.g. Constrained Band Energy, Band Folding, Pyramidal Vector Quantization etc.).

Actually, most of that's in the Specification. If you want to know about Psychoacoustic Models, the general principles are explained in numerous places with reference to other codecs, and most of these techniques are applicable to the tools available in Vorbis. You might also want to look into AoTuV source code versus libvorbis to spot differences in tuning. I think the libvorbis source code is probably pretty good documentation, then search Wikipedia, Google Scholar, and finally Google generally, to understand terms you don't recognise.


Hi Thank you very much for your suggestions..

Regards
Vinod
Go to the top of the page
+Quote Post
testyou
post Jan 28 2013, 08:12
Post #6





Group: Members
Posts: 99
Joined: 24-September 10
Member No.: 84113



QUOTE (kumarsvinod20 @ Jan 25 2013, 01:34) *
downloaded the libvorbis-aotuv_b6.03 as i have many doubts

What do you mean?
Go to the top of the page
+Quote Post
kumarsvinod20
post Jan 28 2013, 09:50
Post #7





Group: Members
Posts: 5
Joined: 25-January 13
Member No.: 106176



QUOTE (testyou @ Jan 28 2013, 12:42) *
QUOTE (kumarsvinod20 @ Jan 25 2013, 01:34) *
downloaded the libvorbis-aotuv_b6.03 as i have many doubts

What do you mean?

I mean , I need some documents for better understanding of the complete compression algorithm used in this code.
Go to the top of the page
+Quote Post
Dynamic
post Jan 28 2013, 17:30
Post #8





Group: Members
Posts: 817
Joined: 17-September 06
Member No.: 35307



Hi Vinod,

I was just checking something out about how SILK works, and was reminded that another good resource is the PDF version of a LibreOffice or OpenOffice Impress presentation which is probably the one used by Jean-Marc Valin in his LinuxConf.au talk "Opus - The Swiss Army Knife Of Audio Codecs". (That PDF was for a gstreamer-related presentation)

This contains links to relevant Wikipedia articles and illustrative graphics. You can also find the video under Presentations on the opus-codec.org website. (The specific presentation materials for that talk are linked in that page and are very similar to the gstreamer link I gave above)

You'll find a good bit of detail including the LP filters coefficients that mirror the vocal tract, the way they're converted to Line Spectral Frequencies before quantization, and the combination of fixed codebooks and scalar and vector quantization, all with helpful diagrams and a fair number of equations and wikipedia links to further info, plus references that you can follow up.
Go to the top of the page
+Quote Post
Dynamic
post Jan 28 2013, 19:06
Post #9





Group: Members
Posts: 817
Joined: 17-September 06
Member No.: 35307




OOPS!


I just realised I posted links to Opus documentation and you want Vorbis. Some of it is relevant (e.g. Pyramid Vector Quantization) but a lot of it is different and reflects the lessons learned in Vorbis development (e.g. Opus/CELT explicitly encodes the band energy, Vorbis indirectly reflects this by the noise curve) and specifics relating to speech coding and low latency that Vorbis doesn't have to worry about.

This post has been edited by Dynamic: Jan 28 2013, 19:06
Go to the top of the page
+Quote Post
db1989
post Jan 28 2013, 19:15
Post #10





Group: Super Moderator
Posts: 5275
Joined: 23-June 06
Member No.: 32180



QUOTE (testyou @ Jan 28 2013, 07:12) *
QUOTE (kumarsvinod20 @ Jan 25 2013, 01:34) *
downloaded the libvorbis-aotuv_b6.03 as i have many doubts
What do you mean?

Iím not aware of the linguistic basis, or how lessons (or is it online translators?) donít seem to correct it very often, but many non-native English speakers use the word doubt when the concept theyíre actually thinking of is question.
Go to the top of the page
+Quote Post
kumarsvinod20
post Jan 29 2013, 05:25
Post #11





Group: Members
Posts: 5
Joined: 25-January 13
Member No.: 106176



QUOTE (Dynamic @ Jan 28 2013, 22:00) *
Hi Vinod,

I was just checking something out about how SILK works, and was reminded that another good resource is the PDF version of a LibreOffice or OpenOffice Impress presentation which is probably the one used by Jean-Marc Valin in his LinuxConf.au talk "Opus - The Swiss Army Knife Of Audio Codecs". (That PDF was for a gstreamer-related presentation)

This contains links to relevant Wikipedia articles and illustrative graphics. You can also find the video under Presentations on the opus-codec.org website. (The specific presentation materials for that talk are linked in that page and are very similar to the gstreamer link I gave above)

You'll find a good bit of detail including the LP filters coefficients that mirror the vocal tract, the way they're converted to Line Spectral Frequencies before quantization, and the combination of fixed codebooks and scalar and vector quantization, all with helpful diagrams and a fair number of equations and wikipedia links to further info, plus references that you can follow up.


Hi, Thank you very much for the information provided above.

Regards
Vinod
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 17th September 2014 - 17:11