Open Source Fraunhofer AAC Encoder (fdk-aac) |
![]() ![]() |
Open Source Fraunhofer AAC Encoder (fdk-aac) |
Aug 27 2012, 00:27
Post
#51
|
|
|
Group: Members Posts: 3 Joined: 26-August 12 Member No.: 102671 |
|
|
|
|
Aug 27 2012, 12:39
Post
#52
|
|
![]() Server Admin Group: Admin Posts: 4810 Joined: 24-September 01 Member No.: 13 |
The thing with apple is pretty clear I guess: they are one of the major patent holders for aac probably (am I right?); I don't think so. They weren't involved in AAC's development at all. QUOTE Honestly, ppl get harassed for posting something that is considered to be illegal by HA. Instead, I think AAC should be considered illegal here instead Some companies licensed encoders and made those free to download. Winamp And Nero are two of them. You're free to discuss those as well as source implementations. |
|
|
|
Aug 27 2012, 12:43
Post
#53
|
|
![]() Server Admin Group: Admin Posts: 4810 Joined: 24-September 01 Member No.: 13 |
This is really off-topic, but I don't think AAC is very succeful too. It have too many different profiles, many of them just eclipse one or another (just see this page here, that describes the entire MPEG-4 Audio; you have two codec for speech, three for general audio and three codecs for lossless). The majority of devices only supports the AAC-LC, many of them didn't support even the AAC Main Profile. The standard is so complex that some codecs never seem a working implementation (like ALS)- I don't see how the fact that AAC (or H264, or any MPEG codec) has a load of cruft and extra profiles that nobody uses has any bearing on how successful it is. People used the parts that made sense and dropped the ones that just existed to satisfy some patent holders or niche applications. Everybody now understands that when you talk about AAC you talk about LC-AAC and it's extensions (HE-AAC, SLS etc). ALS probably never had a working implementation because it was years late to the party and didn't bring any significant advantages. That can't be said for LC-AAC and HE-AAC. The parts that made sense were very successful. There's a hardware decoder in nearly every mobile device out there. |
|
|
|
Aug 27 2012, 12:49
Post
#54
|
|
![]() Server Admin Group: Admin Posts: 4810 Joined: 24-September 01 Member No.: 13 |
QUOTE Overall, all pv or vo codecs that google bought from packetvideo or visual on are clear ripoffs from 3gpp code and they aren't even well done. This was pointed out here a long time ago and reported to Google. I wouldn't be surprised if the sudden appearance of the Fraunhofer encoder in Android was related to it. QUOTE My personal opinion is that they invent new crap not because there is a need, but because they want to prevent from new players coming into the big game. This sentence directly contradicts itself. If a new player with a better algorithm that outperforms the MPEG ones would enter the market, the MPEG would lose out. So obviously there is a need to stay ahead of the game. QUOTE Just the idea that all these companies suddenly collaborate on a publicly available BSD licensed library is really fishy. To me it's clear what their intentions are. I don't think it is a "secret" or "fishy" that the companies who contribute to MPEG standards do so because they expect to license their patents to any implementers of the standard. Or that the public code is crap because they want to license your their software as well. The programmers and researches who contribute expect to get paid, somehow. |
|
|
|
Aug 27 2012, 13:15
Post
#55
|
|
|
Group: Members Posts: 1234 Joined: 5-October 01 Member No.: 220 |
QUOTE My personal opinion is that they invent new crap not because there is a need, but because they want to prevent from new players coming into the big game. This sentence directly contradicts itself. If a new player with a better algorithm that outperforms the MPEG ones would enter the market, the MPEG would lose out. So obviously there is a need to stay ahead of the game. Think he means that the big players pushed AAC to use it as a lock in format that everyone uses and the big corps can control |
|
|
|
Aug 28 2012, 06:21
Post
#56
|
|
|
Group: Members Posts: 34 Joined: 8-August 11 Member No.: 92854 |
Can anyone provide a link to instructions for compiling avconv or ffmpeg with fdk-aac support? Thanks
|
|
|
|
Aug 28 2012, 11:05
Post
#57
|
|
|
Group: Members Posts: 79 Joined: 9-May 10 Member No.: 80499 |
Can anyone provide a link to instructions for compiling avconv or ffmpeg with fdk-aac support? Thanks http://forum.doom9.org/showthread.php?p=1587426#post1587426 http://forum.doom9.org/showthread.php?p=1587513#post1587513 |
|
|
|
Aug 30 2012, 02:12
Post
#58
|
|
![]() Group: Members Posts: 1 Joined: 30-August 12 Member No.: 102755 |
Can anyone provide a link to instructions for compiling avconv or ffmpeg with fdk-aac support? Thanks See Compile FFmpeg on Ubuntu. It contains fdk-aac (from Git) compile instructions. |
|
|
|
Aug 30 2012, 07:14
Post
#59
|
|
![]() Server Admin Group: Admin Posts: 4810 Joined: 24-September 01 Member No.: 13 |
|
|
|
|
Aug 31 2012, 12:16
Post
#60
|
|
|
Group: Members Posts: 46 Joined: 14-June 11 Member No.: 91517 |
I've spent a hour trying to compile this with MSYS/MingW but i gave up eventually because my installation doesn't had the directories the compile instructions mentioned
This post has been edited by darkbyte: Aug 31 2012, 12:27 |
|
|
|
Oct 9 2012, 00:57
Post
#61
|
|
|
Group: Members Posts: 93 Joined: 13-April 07 Member No.: 42452 |
So, this encoder is in the new ffmpeg 1.0. Anyone used it? Just curious.
|
|
|
|
Dec 15 2012, 21:35
Post
#62
|
|
|
Group: Members Posts: 5 Joined: 2-August 12 Member No.: 101956 |
I would like to know too is there someone tested for example: Nero AAC vs Apple AAC vs FDK-AAC
Weird is that FDK is only best open AAC encoder this far, i wonder why someone did not write AAC encoder long time ago like there is Lame that is best for mp3 This post has been edited by jossilint: Dec 15 2012, 21:36 |
|
|
|
Dec 15 2012, 23:08
Post
#63
|
|
|
Group: Members Posts: 437 Joined: 11-February 12 Member No.: 97076 |
I would like to know too is there someone tested for example: Nero AAC vs Apple AAC vs FDK-AAC There are many test. Use the search. QUOTE Weird is that FDK is only best open AAC encoder this far, i wonder why someone did not write AAC encoder long time ago like there is Lame that is best for mp3 FDK-AAC is just a CLI for Fraunhofer, like qaac and qtaacenc are for Apple. This post has been edited by eahm: Dec 15 2012, 23:09 |
|
|
|
Dec 15 2012, 23:30
Post
#64
|
|
|
Group: Members Posts: 1318 Joined: 3-January 05 From: Argentina, Bs As Member No.: 18803 |
|
|
|
|
Dec 15 2012, 23:38
Post
#65
|
|
|
Group: Members Posts: 437 Joined: 11-February 12 Member No.: 97076 |
FDK-AAC is just a CLI for Fraunhofer, like qaac and qtaacenc are for Apple. You can check the OP. My bad, I was confused with fhgaacenc. |
|
|
|
Dec 16 2012, 08:42
Post
#66
|
|
![]() Group: Admin Posts: 4231 Joined: 15-December 02 Member No.: 4082 |
Too bad ffmpeg hasn't sorted gapless encoding yet. It didn't look like avconv supported that yet, either, the last time I looked. Maybe somebody should make a stand-alone encoder.
|
|
|
|
Dec 16 2012, 15:08
Post
#67
|
|
|
Group: Members Posts: 5 Joined: 2-August 12 Member No.: 101956 |
What about quality of FDK compared to closed? is it too some middle quality aac encoder, experts opinion?
Really bad that only closed source encoders have best quality(aac, mp3) So now i know that even Lame is not best quality encoder thanks for info IgorC now i move all my audio to Opus. |
|
|
|
Dec 17 2012, 12:15
Post
#68
|
|
![]() Server Admin Group: Admin Posts: 4810 Joined: 24-September 01 Member No.: 13 |
What about quality of FDK compared to closed? is it too some middle quality aac encoder, experts opinion? Juding from the source code, it should be pretty good: http://www.hydrogenaudio.org/forums/index....st&p=801900 Of course, it should be properly tested. The code might be advanced but full of bugs. (Not likely, but not impossible) |
|
|
|
Dec 21 2012, 07:15
Post
#69
|
|
![]() Group: Developer Posts: 295 Joined: 22-November 10 From: Japan Member No.: 85902 |
Out of curiosity, I tested this codec with various settings using the attached test files and script:
fdkaac_vbr.zip ( 808.61K )
Number of downloads: 145Available VBR qualities for each <profile, #channels, sample rate> by this encoder are as follows: CODE LC-AAC 1ch 11250 1
LC-AAC 1ch 12000 1 LC-AAC 1ch 16000 1,2 LC-AAC 1ch 22500 1,2,3 LC-AAC 1ch 24000 1,2,3,4 LC-AAC 1ch 32000 1,2,3,4 LC-AAC 1ch 44100 1,2,3,4,5 LC-AAC 1ch 48000 1,2,3,4,5 LC-AAC 1ch 64000 1,2,3,4,5 LC-AAC 1ch 88200 1,2,3,4,5 LC-AAC 1ch 96000 1,2,3,4,5 LC-AAC 2ch 8000 1 LC-AAC 2ch 11250 1,2 LC-AAC 2ch 12000 1,2 LC-AAC 2ch 16000 1,2,3 LC-AAC 2ch 22500 1,2,3,4 LC-AAC 2ch 24000 1,2,3,4 LC-AAC 2ch 32000 1,2,3,4,5 LC-AAC 2ch 44100 1,2,3,4,5 LC-AAC 2ch 48000 1,2,3,4,5 LC-AAC 2ch 64000 1,2,3,4 LC-AAC 2ch 88200 1,2,3,4 LC-AAC 2ch 96000 1,2,3,4 LC-AAC 3ch 8000 1 LC-AAC 3ch 11250 1,2 LC-AAC 3ch 12000 1,2 LC-AAC 3ch 16000 1,2,3 LC-AAC 3ch 22500 1,2,3,4 LC-AAC 3ch 24000 1,2,3,4 LC-AAC 3ch 32000 1,2,3,4,5 LC-AAC 3ch 44100 1,2,3,4,5 LC-AAC 3ch 48000 1,2,3,4,5 LC-AAC 3ch 64000 1,2,3,4 LC-AAC 3ch 88200 1,2,3,4 LC-AAC 3ch 96000 1,2,3,4 LC-AAC 4ch 8000 1 LC-AAC 4ch 11250 1,2 LC-AAC 4ch 12000 1,2 LC-AAC 4ch 16000 1,2,3 LC-AAC 4ch 22500 1,2,3,4 LC-AAC 4ch 24000 1,2,3,4 LC-AAC 4ch 32000 1,2,3,4,5 LC-AAC 4ch 44100 1,2,3,4,5 LC-AAC 4ch 48000 1,2,3,4,5 LC-AAC 4ch 64000 1,2,3,4 LC-AAC 4ch 88200 1,2,3,4 LC-AAC 4ch 96000 1,2,3,4 LC-AAC 5ch 8000 1 LC-AAC 5ch 11250 1,2 LC-AAC 5ch 12000 1,2 LC-AAC 5ch 16000 1,2,3 LC-AAC 5ch 22500 1,2,3,4 LC-AAC 5ch 24000 1,2,3,4 LC-AAC 5ch 32000 1,2,3,4,5 LC-AAC 5ch 44100 1,2,3,4,5 LC-AAC 5ch 48000 1,2,3,4,5 LC-AAC 5ch 64000 1,2,3,4 LC-AAC 5ch 88200 1,2,3,4 LC-AAC 5ch 96000 1,2,3,4 LC-AAC 6ch 8000 1 LC-AAC 6ch 11250 1,2 LC-AAC 6ch 12000 1,2 LC-AAC 6ch 16000 1,2,3 LC-AAC 6ch 22500 1,2,3,4 LC-AAC 6ch 24000 1,2,3,4 LC-AAC 6ch 32000 1,2,3,4,5 LC-AAC 6ch 44100 1,2,3,4,5 LC-AAC 6ch 48000 1,2,3,4,5 LC-AAC 6ch 64000 1,2,3,4 LC-AAC 6ch 88200 1,2,3,4 LC-AAC 6ch 96000 1,2,3,4 HE-AAC 1ch 22500 1 HE-AAC 1ch 24000 1 HE-AAC 1ch 32000 1,2 HE-AAC 1ch 44100 1,2,3 HE-AAC 1ch 48000 1,2,3 HE-AAC 1ch 64000 1,2,3,4 HE-AAC 1ch 88200 1,2,3,4,5 HE-AAC 1ch 96000 1,2,3,4,5 HE-AAC 2ch 16000 1 HE-AAC 2ch 22500 1,2 HE-AAC 2ch 24000 1,2 HE-AAC 2ch 32000 1,2,3 HE-AAC 2ch 44100 1,2,3,4 HE-AAC 2ch 48000 1,2,3,4 HE-AAC 2ch 64000 1,2,3,4,5 HE-AAC 2ch 88200 1,2,3,4,5 HE-AAC 2ch 96000 1,2,3,4,5 HE-AAC 3ch 16000 1 HE-AAC 3ch 22500 1,2 HE-AAC 3ch 24000 1,2 HE-AAC 3ch 32000 1,2,3 HE-AAC 3ch 44100 1,2,3 HE-AAC 3ch 48000 1,2,3 HE-AAC 3ch 64000 1,2,3,4,5 HE-AAC 3ch 88200 1,4,5 HE-AAC 3ch 96000 4,5 HE-AAC 4ch 16000 1 HE-AAC 4ch 22500 1,2 HE-AAC 4ch 24000 1,2 HE-AAC 4ch 32000 1,2,3 HE-AAC 4ch 44100 1,2,3 HE-AAC 4ch 48000 1,2,3 HE-AAC 4ch 64000 1,2,3,4,5 HE-AAC 4ch 88200 1,3,4,5 HE-AAC 4ch 96000 3,4,5 HE-AAC 5ch 16000 1 HE-AAC 5ch 22500 1,2 HE-AAC 5ch 24000 1,2 HE-AAC 5ch 32000 1,2,3 HE-AAC 5ch 44100 1,2,3 HE-AAC 5ch 48000 1,2,3 HE-AAC 5ch 64000 1,2,3,4,5 HE-AAC 5ch 88200 1,3,4,5 HE-AAC 5ch 96000 3,4,5 HE-AAC 6ch 16000 1 HE-AAC 6ch 22500 1,2 HE-AAC 6ch 24000 1,2 HE-AAC 6ch 32000 1,2,3 HE-AAC 6ch 44100 1,2,3 HE-AAC 6ch 48000 1,2,3 HE-AAC 6ch 64000 1,2,3,4,5 HE-AAC 6ch 88200 1,4,5 HE-AAC 6ch 96000 4,5 HE-AACv2 2ch 32000 1 HE-AACv2 2ch 44100 1,2 HE-AACv2 2ch 48000 1,2 HE-AACv2 2ch 64000 1,2,3 HE-AACv2 2ch 88200 1,2,3,4 HE-AACv2 2ch 96000 1,2,3,4 This post has been edited by db1989: Mar 1 2013, 18:27
Reason for edit: changing [code] to [codebox]
|
|
|
|
Jan 5 2013, 23:18
Post
#70
|
|
|
Group: Members Posts: 34 Joined: 22-June 12 Member No.: 100900 |
|
|
|
|
Jan 7 2013, 04:58
Post
#71
|
|
![]() Group: Members Posts: 10 Joined: 7-January 13 Member No.: 105656 |
Has anyone tried this frontend yet? Yes, it compiles OK with Ubuntu Linux and cross-compiles OK with mingw. CODE ./fdkaac -m 3 -p 5 -a 1 \ --title "Lucky Night" --album "Treasure Quest Soundtrack" \ --artist "Jody Marie Gnant" --genre "Soundtrack" \ --date "1995" -o Lucky_Night.m4a Lucky_Night.wav QUOTE General Complete name : Lucky_Night.m4a Format : MPEG-4 Format profile : Apple audio with iTunes info Codec ID : M4A File size : 345 KiB Duration : 30s 186ms Overall bit rate mode : Variable Overall bit rate : 93.7 Kbps Album : Treasure Quest Soundtrack Track name : Lucky Night Performer : Jody Marie Gnant Genre : Soundtrack Recorded date : 1995 Encoded date : UTC 2013-01-07 03:47:11 Tagged date : UTC 2013-01-07 03:47:11 Writing application : fdkaac 0.0.5, libfdk-aac 3.3.3, VBR mode 3 Audio ID : 1 Format : AAC Format/Info : Advanced Audio Codec Format profile : HE-AAC / LC Codec ID : 40 Duration : 30s 186ms Bit rate mode : Variable Bit rate : 92.6 Kbps Maximum bit rate : 143 Kbps Channel(s) : 2 channels Channel positions : Front: L R Sampling rate : 44.1 KHz / 22.05 KHz Compression mode : Lossy Stream size : 341 KiB (99%) Encoded date : UTC 2013-01-07 03:47:11 Tagged date : UTC 2013-01-07 03:47:11 |
|
|
|
Mar 1 2013, 04:21
Post
#72
|
|
|
Group: Members Posts: 437 Joined: 11-February 12 Member No.: 97076 |
Thanks nu774, the compiler runs perfectly. Now I have fdkaac.exe, it works well by itself but I can't find the configuration for foobar2000. "-m 4 - %d" doesn't work, "-m 4 %s %d" doesn't work, I can't do only "-m 4 %s".
Do you know a solution? Thanks. This post has been edited by eahm: Mar 1 2013, 04:27 |
|
|
|
Mar 1 2013, 05:40
Post
#73
|
|
![]() Group: Developer Posts: 295 Joined: 22-November 10 From: Japan Member No.: 85902 |
fdkaac requires -o before %d (same as qaac).
|
|
|
|
Mar 1 2013, 06:12
Post
#74
|
|
|
Group: Members Posts: 437 Joined: 11-February 12 Member No.: 97076 |
Of course ...
Thanks. This post has been edited by eahm: Mar 1 2013, 06:12 |
|
|
|
Mar 1 2013, 11:16
Post
#75
|
|
|
Group: Members Posts: 1 Joined: 1-March 13 Member No.: 106961 |
Does anyone knows how to build the fdk acc as a shared library for Android? Thanks!
This post has been edited by db1989: Mar 1 2013, 18:29
Reason for edit: deleting completely pointless full quote of the huge first post
|
|
|
|
![]() ![]() |
|
Lo-Fi Version | Time is now: 19th June 2013 - 00:19 |