• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Tagging AAC Audio

svet-am

Supreme [H]ardness
Joined
Jan 6, 2003
Messages
5,146
I've been doing some work lately (for work) that involved frequency analysis and trade-offs of different codecs on audio quality of different genres of music and different types of audio.

One of the side-effects is that my data has convinced me that AAC is _definitely_ the way to go for the type of music that I listen to. I've believed this for a while, but now I have numbers to back it up.

One thing that concerns me about AAC is the tagging format. I am very well versed in ID3 and how it works. Further, I have some handy-dandy tools that let me customize my ID3 tags just how I want them.

I've been using ID3-TagIt to manage MP3 ID3 tags in the past. Are there any similar apps for AAC tags?

What format does AAC even use? Looking at the spec sheet for the AAC format, it looks like it doesn't actually specify a tagging format. It looks like it's just got space in the container for an arbitrary tagging format.
 
got some of my answers by stalking other forums. The user data blocks in the AAC container is the only thing officially specified. Apple has "standardized" the block types and ordering for the user data section. However, they haven' t released a spec to this so everything known about AAC tagging is just reverse-engineered from Apples implementation.

That really sucks. It really shoots in the foot the usefulness of AAC, regardless of its audio merits.
 
Frequency analysis alone is virtually useless as far as perceived audio quality is concerned. Since every modern format uses a psychoacoustic model, the only way to really test them is to perform listening tests.

Though I don't think your conclusion is a bad one, most tests seem to indicate that a good AAC encoder or Vorbis is the way to go.

There's no official spec for AAC tagging, so virtually every uses Apple's iTunes standard, which isn't really a standard at all since the details are unpublished by Apple. Lots of apps can use it though, I like foobar2000 for tagging, though some other popular ones support AAC as well, like Mp3tag.
 
Frequency analysis alone is virtually useless as far as perceived audio quality is concerned. Since every modern format uses a psychoacoustic model, the only way to really test them is to perform listening tests.
.

It _is_ pretty conclusive when you are considering genres with lots of highs and MP3 just plain ole drops everything over ~19kHz
 
It _is_ pretty conclusive when you are considering genres with lots of highs and MP3 just plain ole drops everything over ~19kHz

Not really, those frequencies are barely audible when they are alone, discarding them makes perfect sense in most psychoacoustic models and most real-world situations since you won't hear the difference. If the remaining audio sounds better, the codec is performing better, regardless of how the spectral looks. Storing that high frequency information takes space in the file, and in most low to medium bitrate situations those bits are better used in the far more critical midrange, which you can't tell from looking at spectrals.

Anyway, AAC and Ogg do sound better, it's just naive to look at spectrals alone. It's not difficult at all to generate a file that cuts off at say 19KHz that sounds much worse than one at an equivalent bitrate cutting off at 16KHz, which was my point.
 
Back
Top