AudioMasters
 
  User Info & Key Stats   
Welcome, Guest. Please login or register.

Login with username, password and session length
May 20, 2010, 01:43:38 AM
70515 Posts in 7368 Topics by 2192 Members
Latest Member: MeetPlanB
News:       Buy Adobe Audition:
+  AudioMasters
|-+  Audio Related
| |-+  General Audio
| | |-+  HE-AAC v2
  « previous next »
Pages: [1] Print
Author
Topic: HE-AAC v2  (Read 3346 times)
« on: May 01, 2008, 11:59:24 AM »
Randall Offline
New Member
*
Posts: 2



Hello guys, this is my first time to join a forum.  I have read the topics that were posted here and I think I would really learn a lot from this forum.
Curruntly I'm using WMA 10/(9.1 Prof):
Bit Depth: 24-bit
Sample Rate: 48 khz
Channels: Stereo
Bit Rate: (VBR) Quality 25

But lately I found out that HE-AAC v2 sounds better that what i'm using right now at lower bit rates.  But I'm really new at HE-AAC v2.  Could you please tell me what bit rate, bit depth, sample rate, channels I should use to store my audio files at near/ cd quality.  And is there a plugin that can help me encode freely using m4a?  Right now I have downloaded a plugin from rarewares/aac.com (I thnk, but I'm not sure if thats the right address) but its very limited in terms of what I options I can set.  How can I convert my audio files to HE-AAC v2 using AA2/3?
And what is the difference between (or which is better) CBR,VBR,ABR?
Logged
Reply #1
« on: May 01, 2008, 12:06:41 PM »
pwhodges Offline
Member
*****
Posts: 1125

WWW

I don't know about plugins, but the free Nero command-line encoder here is widely recommended.  I wouldn't use the HE versions of the encoding, as I have found programs that don't yet play them.

But for moderate bit rates, ogg/vorbis is often said to be the best.

Paul
Logged
Reply #2
« on: May 01, 2008, 08:33:37 PM »
AndyH Offline
Member
*****
Posts: 1606



The goal is often called transparency, but under whatever label, it means that you are unable to tell which is the unencoded original and which is the lossy encoded version. The way you determine what is transparent is through ABX testing, which is very easily done on a computer.

A number of freeware programs will randomly present one or the other to you. You select which you believe to be the original. After a number of trials, the program shows you your score and the probability that any correct choices were other than mere guesses. When you can’t tell the difference, you have an adequate format. There is a basic explanation here
http://www.hydrogenaudio.org/forums/index.php?showtopic=16295&
There is more in-depth information on the PCABX site and the Hydrogen audio site.

Several programs are available
http://www.pcabx.com/
http://www.kikeg.arrakis.es/winabx/
http://ff123.net/abchr/abchr.html
Foobar2000 is also often used with an ABX plugin
http://www.foobar2000.org/
Logged
Reply #3
« on: May 02, 2008, 12:09:31 PM »
Randall Offline
New Member
*
Posts: 2



So it really depends on what you think is good enough/more likely what sounded like the original one.  I'm planing on using this set of specs. on my auido files, do you have any suggestion or comments?
Specs:
Bit Depth: 16 bit
Sample Rate: 48khz
Channels: Original
Bit Rate: VBR Quality 15 (estimated 32)

And what is the difference and advantages between CBR/VBR/ABR?
Logged
Reply #4
« on: May 02, 2008, 12:20:39 PM »
pwhodges Offline
Member
*****
Posts: 1125

WWW

What is transparent to you can only be determined by your own tests (see here).  However, you can get a clue for starting points by looking at tests that have already been done, such as this one (see here for a useful list).  Note that the main codecs are not very clearly differentiated in performance these days; there's only so much that one can throw out of a signal before there is insufficient to fool the ear, and IME we are unlikely to see much further change in that however sophisticated the codecs get (we are already throwing out 9/10* and keeping remarkable fidelity - but note that this is only for listening; archiving or further processing is another matter altogether).

Paul

* Given that lossless encoding typically halves the amount of data, we could argue that only 4/5 of the data is being thrown out by the perceptual encoding, which is still remarkable.
Logged
Pages: [1] Print 
« previous next »
Jump to:  

Powered by MySQL Powered by PHP Valid XHTML 1.0! Valid CSS! Ig-Oh Theme by koni.