Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Should I resample to 48kHz during converting to AAC (Opus-like)? (Read 1908 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Should I resample to 48kHz during converting to AAC (Opus-like)?

I am going to encode my lossless collection of ripped CDs into 144 kbps AAC (QAAC) for listening on both Android phone and Windows (Foobar2k with an onboard Realtek DAC and good speakers/headphones). My goal is to keep it lossy, but make sure I retain enough quality to keep the music transparent for my ears. Should I include 44.1 -> 48 kHz resampler in the Foobar conversion chain to get 16b/48kHz output AAC files? The resampling practically does not affect the conversion speed and resulting file size. The benefit as I can see it will be that my AAC files will match the native sampling rate of my phone/PC DACs, which is as far as I know 48 kHz, thus no need for software resampling during playback - better sound quality and less battery drain.

Such approach was implemented in Opus, and I cannot think of any downsides of it. Are there any? Am I correct in my assumptions? Thanks in advance for the replies.

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #1
Quote
thus no need for software resampling during playback - better sound quality and less battery drain.
My assumption would be, no difference...  so probably no harm...

Quote
does not affect the conversion speed and resulting file size.
The file size is given by the bitrate - 144kbps = 144 kilobits per second.   And if you know there are 8 bits in a byte, 18 kilobytes per second, or 1080kB per minute.   (Maybe plus some space for any embedded album artwork and/or other metadata.)

Quote
16b/48kHz output AAC files?
There is a sample rate but AAC (and MP3, etc.) don't save individual samples so there is no bit depth.


Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #2
Will changing sample rate from 44.1 to 48 kHz affect the efficiency of conversion (sound quality) at given bitrate?

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #3
Nope, but I would stay at 44.1 kHz.
Most ABX tests were performed 44.1 kHz.
qaac is Apple encoder and Apple sells 44.1 kHz music so it is safe to assume that most tunings were done at 44.1 kHz.
gold plated toslink fan

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #4
Thank you.

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #5
Probably what I need to do is to try to find out if I can hear any difference between 44.1 and 48 kHz AAC files on my hardware.

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #6
Indeed, that is what you must try.
It looks like most Android phones run the DAC at 48 kHz (Video!).
As a consequence all 44.1 audio must be resampled to 48.
It is thinkable that one of the re-samplers (Foobar or Android) does a better job but today re-sampling is most of the time pretty transparant.
TheWellTemperedComputer.com

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #7
Quote
Probably what I need to do is to try to find out if I can hear any difference between 44.1 and 48 kHz AAC files on my hardware
You can try a quick-casual listening test and if you don't hear a difference that's what I'd expect.      But it's easy to fool yourself so if you think you are hearing a difference I'd recommend a blind ABX test to see if you get statistically significant results.   I don't know if there's ABX software for Android so you might need someone else to make some "randomized" files to keep the test blind.

You also mentioned battery life so you could experiment with that too.    But there may be some unknown variables so that test should probably be repeated multiple times to confirm consistent-repeatable results.    (Or if there's not much difference the 1st time, or if the battery lasts longer at 44.4kHz on the 1st test, you can probably stop.)

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #8
Personally, I really doubt you could tell the difference and if anything you'd make things slightly worse by doing that.

The reason Opus does the 48 khz thing is to make the codec slightly more simple, I believe.

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #9
I do resample even for Vorbis encoding, so when I play it, it will not be resampled with some primitive stuff. AFAIK most HW runs at 48 kHz.

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #10
Audibility? Nah. Battery life? That could be tested. Make a "48000" playlist and a "44100" playlist of the same tracks, put on repeat, observe battery life..

I know it is a tough test to perform, given that one has to put the phone down for too long - twice! ;-)
High Voltage socket-nose-avatar

 

Re: Should I resample to 48kHz during converting to AAC (Opus-like)?

Reply #11
I need to do is to try to find out if I can hear any difference between 44.1 and 48 kHz AAC files on my hardware.
I doubt you could find the different

I read here, the writer notice he prefers don't increase re-sample cause interpolation, it recreates waveform using stored number might possible
https://www.psaudio.com/copper/article/sample-rate-conversion/

then the next article said increasing sample rate could help when the increasing bit-depth
https://www.izotope.com/en/learn/digital-audio-basics-sample-rate-and-bit-depth.html

yep, that's complicated, as far I know modern days technologies continue to evolve, digital audio processing get matured
re-sampling 44.1KHz from Lossless to 48KHz lossy format it's not a problem to worry about, except from lossy to lossy  :))

Opus 96-192Kbps and Aac-lc 96-128Kbps
Source 32bit floating point is fine don't need dithering (dB noise -758dBFS that's a lot)
Source Fixed Point aka Integer should use dithering to prevent truncation distortion
Source is RAW

 
SimplePortal 1.0.0 RC1 © 2008-2021