Skip to main content
JohnVo
Inspiring
July 12, 2017
Answered

increasing sample rate does improve the audio restauration?

  • July 12, 2017
  • 2 replies
  • 578 views

Hi

i know audition cc work in 32bit float ,but may i ask increasing the sample rate from 44100 to an higher value does it improve audio rastauration ?

i mean will this give to audition more data to elaborate better the audio?

thanks

This topic has been closed for replies.
Correct answer ryclark

Not really since anything above the human hearing threshold won't make any difference to the end result. This is especially true if you are doing restoration of audio that is not of the best quality to start with ie. audio cassette tapes. All you will do is to make your files larger. The bit depth is important though since this is where quality can get lost if you are doing lots of processing of the audio signal. So when digitising always use a high quality audio interface which will work at 24 bits and store after ingesting with Audition at 32 bit float.

2 replies

SteveG_AudioMasters_
Community Expert
Community Expert
July 12, 2017

Giovannivolontè  wrote

i mean will this give to audition more data to elaborate better the audio?

As ryclark says, it won't. All you will be doing is providing more noise for Audition to work on. Generally this is not helpful at all; you are just making the software work harder on things it doesn't need to be doing, for no audible benefit whatsoever.

JohnVo
JohnVoAuthor
Inspiring
July 12, 2017

hi

and may i know when should I use higher sample rate ?

i can see i can use even 192000....

thanks

SteveG_AudioMasters_
Community Expert
Community Expert
July 12, 2017

Giovannivolontè  wrote

and may i know when should I use higher sample rate ?

i can see i can use even 192000....

For audio, there is no reason to use anything above 48k at all - and only up to 48k because it's standard for video work - there is no audible benefit whatsoever. The only people who need to sample at any rates like that would be the ones analysing bat sounds. The rules are very simple; you can represent all frequencies at all levels up to the Nyquist frequency for any sample rate. The limit of human hearing (and only when you're a child) is 20kHz, so nothing higher than that needs to be recorded or processed - which means that 44.1k is sufficient for all normal audio.

It used to be the case that because of the brick-wall anti-alias filters used in some ADCs, the sound coming back was of a generally less 'phasey' quality if you used 88.2 or 96k sample rates, but all modern-day converters use oversampling techniques to get around that issue completely, so there is no difference in the sound you hear whatsoever. The difference wasn't anything to do with the rate itself, but it meant that the filter could be operated way outside the audio band, so no phase effects within it occurred.

Several years ago, some careful academic research by a noted academic was carried out into whether anybody could tell the difference between a CD (44.1k 16-bit) and any of the higher resolution formats. The research was duplicated using a lot of volunteers, and several different high-class reproduction systems. The conclusion? That it was not possible for humans to tell the difference at all.

ryclark
ryclarkCorrect answer
Participating Frequently
July 12, 2017

Not really since anything above the human hearing threshold won't make any difference to the end result. This is especially true if you are doing restoration of audio that is not of the best quality to start with ie. audio cassette tapes. All you will do is to make your files larger. The bit depth is important though since this is where quality can get lost if you are doing lots of processing of the audio signal. So when digitising always use a high quality audio interface which will work at 24 bits and store after ingesting with Audition at 32 bit float.