How to convert FLAC to AAC

[ILoveEDM]

From what I know of FLAC is supposed to be the best audio format but since it's not supported on all devices I want to convert it to AAC. I want a high quality conversion and I downloaded LameXP. I also read that VBR and ABR sometimes cause problems playing so I guess I have to stick to CBR even though VBR is supposed to give you the highest quality. What I was thinking then is dragging the arrow all the way to 500 kbps and then clicking Encode Now. Is there anything in the Advanced Options I should change or am I set?

[hello_hello]

I don't use LameXP but does it have a quality setting for the Nero AAC encoder? I use the default quality of 0.50 when converting to AAC and I've not had any problems playing the converted audio with a device which supports AAC (TV/Bluray player etc).... even though it's variable bitrate. If the resulting output is too small/large for you then lower/increase the quality setting.
I've read the same thing regarding VBR MP3 too, but I've never had a problem playing VBR MP3 with a device which supports MP3. Don't believe everything you read.

If LameXP doesn't have a quality setting for encoding with NeroAAC there's plenty of conversion GUIs which do. TAudioConverter for example. There's no need to play with any other encoder settings. Just pick the quality.

[ILoveEDM]

Well in the Advanced Options it says LAME Algorithm Quality and it's set to the 3rd line that says High Quality (Recommended). Idk if that's what you mean though but yeah it does use the Nero AAC Encoder.

[hello_hello]

I downloaded LameXP for a look.....

Lame is the MP3 encoder. The Lame algorithm quality tells Lame which algorithm to use when it's encoding using a constant bitrate. The bitrate can still be specified separately, while the algorithm effects the encoding speed. The default algorithm is 3. Most programs seem to use 2 as the quality is better (which I'd guess LameXP does). 1 tends to be very slow.
If you select VBR MP3 encoding under the Compression tab, the algorithm setting would have no effect (I'm fairly sure it only applies to CBR encoding). "Quality Level 2" is (I'm pretty sure) the default Lame VBR quality.

If you switch to the NeroAAC encoder and select VBR encoding the Quality slider changes to display the NeroAAC way of doing things. I'm pretty sure "Quality Level 0.50" is the default for Nero although LameXP seems to use Q0.45 by default for some reason. Anyway, lower quality settings will produce smaller file sizes etc. I just use Q0.50 for everything myself.

[noemi7]

AAC is acoustically horrible codec. Even on a low HiFi system you will notice that in comparison with original CD quality audio it lacks lower frequencies, and sounds flat. So it's better imo to convert audio to MP3, or if your system supports them: AC3 or MP2. FLAC can be muxed into MKV.

[Hoser Rob]

AAC is acoustically horrible codec.

I have to disagree with this. It actually has better resolution, and I've seen some pretty convincing technical reasons to support that.

Here's one: http://www.stereophile.com/features/308mp3cd/

However, the difference is pretty subtle.

I wonder whether the aac files you're talking about are standard iTunes downloads. Those are pretty awful. The settings they use are not good for hifi.

So it's better imo to convert audio to MP3.

I do agree with this, but for reasons of device compatibility, which is what the OP was looking for.

I haven't used lamexp, and I won't be because I use another LAME based encoder in linux and lamexp is a windows program. LAME is definitely the best mp3 encoder, but lamexp needs a nero plugin for aac. No nero product will ever touch my one remaining windows partition. In windows I'd probably use foobar (terrible interface but very powerful) or audacity.

I've also heard that stuff about vbr being higher quality than cbr. It ain't true, at least sonically. It's a bit saving measure, not a sonic enhancer. I have been around computers around far too long to trust a lossy encoding algorithm to know when it can shave bits with no noticeable effect. I think that's a big reason why you can't just use one set of video encoding parameters for any video.

I've taken wav/flac files and converted them to 320k mp3 in both modes and then a/b'ed them on my stereo. There is absolutely no doubt in my mind that cbr sounds better.

[hello_hello]

Every lossy audio encoder pretty much has a point where the bitrate is high enough for the encoded version to be "transparent". AAC is no exception. Which AAC encoder do you use noemi7 and what sort of bitrate/quality setting? And no two encoders are created equally. You could encode the same file at the same bitrate with two different MP3 encoders and get two different results.

The problem with MP3 is multichannel audio isn't supported by most MP3 encoders/decoders, whereas it's pretty well supported for AAC. The general consensus seems to be AAC is better than MP3 at low bitrates (under 100kbps) and basically transparent at around the 140kbps point.

Here's a table of NeroAAC quality settings and the corresponding bitrates to expect on average when encoding stereo music.
http://www.hydrogenaudio.org/forums/index.php?showtopic=44310

[hello_hello]

I've also heard that stuff about vbr being higher quality than cbr. It ain't true, at least sonically. It's a bit saving measure, not a sonic enhancer. I have been around computers around far too long to trust a lossy encoding algorithm to know when it can shave bits with no noticeable effect. I think that's a big reason why you can't just use one set of video encoding parameters for any video.

I've taken wav/flac files and converted them to 320k mp3 in both modes and then a/b'ed them on my stereo. There is absolutely no doubt in my mind that cbr sounds better.

The golden eared folks at hydrogenaudio seem to disagree with you on that one. The Lame encoder can use faster or slower algorithms at the same constant bitrate. Do you use the slowest algorithm for CBR encoding?

Here's what hydrogenaudio say about 320kbps MP3 and VBR encoding:
http://wiki.hydrogenaudio.org/index.php?title=LAME

"Maximum quality is achieved when, regardless of listening conditions, you are unable to detect a difference between the MP3 and the original. As demonstrated by blind ABX tests, LAME-encoded MP3s typically achieve this level of transparency when encoded with the default settings, at bitrates well below maximum. Encoding with other settings will have no effect on the quality."

"-b 320 is an alternative to the VBR settings above.
This CBR mode will maximize the MP3's bitrate and overall file size. The extra space may allow for some parts of the audio to be compressed with fewer sacrifices, but to date, no one has produced ABX test results demonstrating that perceived quality is ever better than the highest VBR profiles described above."

Unless you carried out proper ABX test there's no doubt in my mind you've probably shown the benefit of the placebo effect.

I take it you don't use VBR encoding for video then? What bitrate do you use for CBR video encoding which looks transparent for all video? Or do you have more faith in video encoders?

http://www.stereophile.com/content/mp3-vs-aac-vs-flac-vs-cd-page-2
"Both MP3 and AAC introduce fairly large changes in the measured spectra, even at the highest rate of 320kbps."

That sums up why I don't think the article should be taken too seriously. Lossy encoders modify the audio and throw stuff away etc. That's how they work. Unless you listen to music with your eyes, it's what you can hear which matters, not what might be shown in a graph. If you can't hear the difference then the encoder is doing exactly what it's supposed to do. I only skimmed through the article, but I don't think the author ever discussed differences he could hear, only ones he could see.

[El Heggunte]

From what I know of FLAC is supposed to be the best audio format but since it's not supported on all devices I want to convert it to AAC. I want a high quality conversion and I downloaded LameXP. I also read that VBR and ABR sometimes cause problems playing so I guess I have to stick to CBR even though VBR is supposed to give you the highest quality. What I was thinking then is dragging the arrow all the way to 500 kbps and then clicking Encode Now. Is there anything in the Advanced Options I should change or am I set?

Well, I don't use LameXP, but anyway, AAC at 500kbps is superoverkill for stereo music at 44.1kHz. Many/most people say stereo AAC is perceptually-transparent at 128kbps if you use a "5-star" encoder. NeroAacEnc still is very-good, however many people say Apple's AAC Compressor "has always been" the best one You can also give a try to Fraunhofer's DLL, available through the recent builds of Winamp. Just for the notes, I prefer and recommend anything between 160kbps and 192kbps, as long the storage space is not a fundamental factor of course

[dfisher052]

If you're looking for just a simple FLAC to AAC converter try Freemake Audio Converter. It works for me.

[Hikmet]

Use "AudioCoder" and 192 kbps enough for stereo. Up this bitrate necessary only specific operations.

[Hoser Rob]

I've also heard that stuff about vbr being higher quality than cbr. It ain't true, at least sonically. It's a bit saving measure, not a sonic enhancer. I have been around computers around far too long to trust a lossy encoding algorithm to know when it can shave bits with no noticeable effect. I think that's a big reason why you can't just use one set of video encoding parameters for any video.

I've taken wav/flac files and converted them to 320k mp3 in both modes and then a/b'ed them on my stereo. There is absolutely no doubt in my mind that cbr sounds better.

The golden eared folks at hydrogenaudio seem to disagree with you on that one. The Lame encoder can use faster or slower algorithms at the same constant bitrate. Do you use the slowest algorithm for CBR encoding?

Here's what hydrogenaudio say about 320kbps MP3 and VBR encoding:
http://wiki.hydrogenaudio.org/index.php?title=LAME

"Maximum quality is achieved when, regardless of listening conditions, you are unable to detect a difference between the MP3 and the original..."

Well, that just proves they may be good computer geeks but they're not exactly "golden eared".

I agree that with some source material ... ie. really badly recorded stuff or many old recodings or most popular recordings of the last decade that are so over compressed there's no low level information to lose ... and on lower quality playback gear you can't tell the difference between a well done mp3 and the wav source.

But on any decent recording, with a decent playback system, that claim is just nonsense.

However, many people are use computer speakers to listen to music nowadays and mp3s or aacs are just fine for them. I don't have any problem with that.

[hello_hello]

Well, that just proves they may be good computer geeks but they're not exactly "golden eared".

I might need you to explain how it proves that. I don't see the logic there. The Lame developers don't seem to agree with your conclusion. The Lame site has a link to the hydrogenaudio page as a list of recommended settings.

I agree that with some source material ... ie. really badly recorded stuff or many old recodings or most popular recordings of the last decade that are so over compressed there's no low level information to lose ... and on lower quality playback gear you can't tell the difference between a well done mp3 and the wav source.

So which recordings does that leave as ones which have low level information? You seem to have eliminated most of them.

If you'd care to encode one of those decent recordings using VBR and CBR and upload both for others to listen to, I'd be keen to hear the difference myself. Or upload a sample of the original audio for others to encode themselves.
And CBR MP3 isn't exactly CBR anyway. The Lame encoder uses a "bit reservoir" which allows it to save unused bits in one frame for use in the next one. Unused bits are saved as "padding" to keep the bitrate constant and possibly giving the impression the algorithm isn't shaving bits in CBR mode and reducing the quality as a result. MP3packer claims to be able to reduce the size of a CBR, 320kb/s MP3 up to 10% by removing the padding and converting the CBR MP3 to VBR. It'll also re-arrange the bits in a VBR MP3 to make it CBR.
http://wiki.hydrogenaudio.org/index.php?title=CBR

But on any decent recording, with a decent playback system, that claim is just nonsense.

I assume you've just listened to a VBR MP3 and then a CBR MP3 and heard a difference? Was it at least an MP3 encode of the same track? Which VBR preset was used? To be honest in the absence of a proper ABX test I take any claims which rely on human perception with a grain of salt.

[ozok]

As hello_hello said, you can try TAudioConverter. It can use QAAC, FFMpeg, Fraunhofer's encoder, fdk-aac or neroaac. QAAC needs quicktime or apple software support, Fraunhofer needs winamp to be installed. FDK-AAC can be used without installing anything but it is said to perform worse quality wise amongst that have support for AAC-HE. FFMpeg doesn't have HE support and general quality is the worst among the lot. Nero is a decent encoder but hasn't been updated in a while. Also you must place it and neroaactag.exe next to TAC.exe manually.

I personally prefer qaac. If you ask me the difference isn't that much if you go above a certain bitrate.

FDK-AAC might be worth to take a look if you don't want to install anything.