HydrogenAudio

Hydrogenaudio Forum => General Audio => Topic started by: deathcoreRULES on 2016-05-11 23:56:09

Title: Finding the true bitrate of an audio file with spectrogram
Post by: deathcoreRULES on 2016-05-11 23:56:09
Hi everyone, it's been a while since I have visited the forum.

The other day I was reading a post on reddit where a user said you should always check the albums/songs you buy off iTunes or other music websites to verify they are the bitrate they claim to be.  Now I don't really have any doubts about the companies selling the music, but I was curious anyway, so I downloaded Spek.

From another reddit post I found this.
Code: [Select]
*MP3 file, Bitrate 64 kbps. Cut-off at 11kHz.
*MP3 file, Bitrate 128 kbps. Cut-off at 16 kHz.
*MP3 file, Bitrate 192 kbps. Cut-off at 19 kHz.
*MP3 file, Bitrate 320 kbps. Cut-off at 20 kHz.
*M4A file, Bitrate 500 kbps. Cut-off at 22 kHz.
*FLAC file, Lossless quality (Bitrate usually 1000 kbps or higher). Graph's drawn continuously, no cut-off.

(http://i.imgur.com/ZLr9WDK.png)
The list above makes sense, but I am confused on two things, so I uploaded a photo.  My first picture is of Time, the Pink Floyd song.  I ripped the album maybe 10 years ago with iTunes.  It's 128 kbps, but it's AAC, not MP3.  It is my understanding that AAC at lower bitrates, sounds better than MP3 at equivalent or even slightly higher bitrates.  Am I wrong?  Spek shows it pretty much 18 kHz across; A 128 kbps MP3 should cut off at 16 kHz.

(http://i.imgur.com/HqUPIj7.png)
My second picture is of Kanye West's song We Don't Care.  At the top it says the file is 128 kbps MP3, but it cuts off at 14 kHz.  So can I assume this is actually like 96 kbps or around there?  Though one thing throws me off...  What are those dark purple and dark blue spikes that hit 22 kHz?  The reddit guy who posted this, says that is just from a poor encoder and that any audio file should have it's kHz go straight across, no little spikes or peaks.
Title: Re: Finding the true bitrate of an audio file with spectrogram
Post by: greynol on 2016-05-12 01:22:39
You can't judge bitrate (or quality for that matter) based on a spectrogram.  You determine bitrate from the size of the audio data and its duration.

Re: "straight accross"
The reddit guy doesn't know what he's talking about. I won't be bothered to read anything else he has to say on the subject.

The spikes all the way to the top are likely the result of clipping or some other discontinuity in the signal.
Title: Re: Finding the true bitrate of an audio file with spectrogram
Post by: greynol on 2016-05-12 01:24:48
PS:
This very topic was covered recently. Did you not find it when searching the forum for an answer?
Title: Re: Finding the true bitrate of an audio file with spectrogram
Post by: sven_Bent on 2016-05-12 02:34:19
Just logged in to second greynol.  There is not set in stone relation  between cut off and quality or bitrate.
I believe its pretty much explained in the hydrogen audio wiki ( i might be wrong).

You can easily make a 128kbits mp3 with 20khz cutoff and it might even sound worse than a 112kits with a 16khz cut-off
One pieces of music might have lots of artifacts at 192kbits  another could sound perfectly transparant at 128kbit
Title: Re: Finding the true bitrate of an audio file with spectrogram
Post by: greynol on 2016-05-12 03:39:30
https://hydrogenaud.io/index.php/topic,111736.msg920832.html
SimplePortal 1.0.0 RC1 © 2008-2021