-
1. Re: How to get the best vocals out of this poor recording?
emmrecs Dec 25, 2011 1:30 AM (in response to Srivas 108)Hi.
Short answer to your question, IMO: not very much!
The file has a 16kHz sample rate so there is no frequency aboe 8 kHz, admittedly not "vital" to speech only, but it does mean the sound has no "air" or real clarity to it. You cannot, as you've found, add High Frequency ("treble") and effect much change. In addition it sounds as if the two voices are "going on and off" mic very quickly; there's almost a "phasing" sound. Perhaps the original recording was made on a machine where "Automatic Record Level" or "Automatic Gain Control" was in operation and so produces the noticeable "pumping" effect to the sound.
As I said, this is all my opinion. There are a number of users of this forum who are much more experienced than me who may well be able to offer much more help to you.
Jeff
-
2. Re: How to get the best vocals out of this poor recording?
topomorto Dec 25, 2011 2:43 AM (in response to emmrecs)The file has a 16kHz sample rate so there is no frequency above 8 kHz
There's even less than that I think - if you look at the file in spectral view, there's nothing above 4k, so it looks like the recording was originally done at an 8kHz sample rate; and in fact there's not much info above 2.5k.
You can take out some of the background noise with an expander curve:
(I'm sure you can do better than that!)
Apart from that, you can see what can be done with eq; unfortunately, vocal enhancer just brings out the distortion....
-
3. Re: How to get the best vocals out of this poor recording?
Srivas 108 Dec 27, 2011 8:26 PM (in response to Srivas 108)Thanks for the help emmrecs and topomorto!
Expander curve did a nice job with the noise, that's good enough for this recording! I've been playing around with noise and hiss reductions a lot, but Dynamics Processing is always a puzzle for me, would be nice to have a hold on it, especially for attenuating specific freq ranges, is there some guide for using it?
I'm trying parametric equalizer for getting more clarity into recording, it sounds better, but again I'm not an expert on that, maybe someone could try and get better results.
Another way is to attenuate the whole thing, use simple equalizer to raise treble and then Hard Limiting for raising the whole volume. Is that an intelligent approach or not?
-
4. Re: How to get the best vocals out of this poor recording?
topomorto Dec 28, 2011 4:22 AM (in response to Srivas 108)I'm not aware of there being an expander with frequency - specific controls. You could always split the recording out by frequency band into two files and process them separately.
As you can see from the spectral view, there's nothing above 4k at all, so whatever you do with Eq, nothing above 4k will make a difference.
You might find that boosting between 2 and 4k gives you some more clarity at the expense of a harsher sound. There's also a bit of resonance around 100hz that you can cut out:
-
5. Re: How to get the best vocals out of this poor recording?
Srivas 108 Dec 28, 2011 6:53 AM (in response to topomorto)Hey Topomorto, that's pretty cool stuff!
I mean probably nothing special to you but I got a bit smarter how to approach things like that. So that's pretty much it, isn't it? Well that's good enough.
topomorto wrote:
You could always split the recording out by frequency band into two files and process them separately.
What do you mean by that?
And one more thing. What do you think, is it reasonable to use Dynamic processing to amplify the lower frequencies without touching the higher ones, to make the whole track more balanced and also louder? Or if I just use Hard Limiter I get the same out of it? I guess Dynamic processing is more precise, but what should be the best settings for doing that? I would like to amplify the whole track like 20 db to make it more even with my other recordings, but avoid clipping and also unnecessaily amplifying the noisy lower frequencies. (I have to figure out how this thing works, I mean dynamic processing, it's a wonderful tool for things like that.)
Thanks for the help!
-
6. Re: How to get the best vocals out of this poor recording?
topomorto Dec 28, 2011 7:25 AM (in response to Srivas 108)topomorto wrote:
You could always split the recording out by frequency band into two files and process them separately.
What do you mean by that?
If you want to do dynamics processing with different settings on different frequency ranges, you either need a multi-band processor, or you need to separate out the frequency ranges into different files.
In Audition, There is a multi-band compressor, but not a multi-band expander.
So, to do expansion/gating, you could take two copies of the file playing simultaneously in multitrack mode, delete or filter the high frequencies from one and the low frequencies from the other, and then process them differently.
And one more thing. What do you think, is it reasonable to use Dynamic processing to amplify the lower frequencies without touching the higher ones, to make the whole track more balanced and also louder?
You could try using the Multiband Compressor to do that job. (that is a multi-band processor, so no need to split up your file in the way I was describing).
That's what modern pop producers do to tracks to maximise the loudness - some people really hate the effect as it can lead to very harsh sounding audio!
By the way, did you know that in the spectral view, you can just select frequency ranges and boost / cut them directly? :
-
7. Re: How to get the best vocals out of this poor recording?
topomorto Dec 28, 2011 7:38 AM (in response to Srivas 108)Dynamics Processing is always a puzzle for me, would be nice to have a hold on it, especially for attenuating specific freq ranges, is there some guide for using it?
'Dynamics Processing' is not a multi-band effect, so in simple terms it won't be able to specifically target specific frequency ranges.
It can act as multiple expanders and/or compressors chanined together. Each dot on the graph is the equivalent of one traditional compressor or expander.
You will see how some of the graphs in this article relate to the interface of 'Dynamics Processing' :
-
8. Re: How to get the best vocals out of this poor recording?
TheJackAttack Dec 28, 2011 9:31 AM (in response to topomorto)A couple of youtubes that sort of help to aurally illustrate dynamic processing....
I don't own these or necessarily advocate them over any other device but the theory and application apply to all dynamics processors.
Jack
-
9. Re: How to get the best vocals out of this poor recording?
ryclark Dec 28, 2011 9:39 AM (in response to TheJackAttack)In the Settings tab of the Dynamic Processing you can filter your audio by adjusting Band Limiting so that if you are using the Expander, for instance, it only works on the selected band of audio. See the DeEsser presets for an example.
-
10. Re: How to get the best vocals out of this poor recording?
Srivas 108 Dec 31, 2011 7:22 AM (in response to Srivas 108)Thank you everyone for helping out. I learned a lot new from this experience. Multiband compressor did a good job, I choosed Raise Vocals and amplified midtones even more, I'm pretty happy with the results, can't get much better than this.
Also started to understand what is compressing and expanding, Wiki has a nice tutorial.
I couldn't use amplification in spectral view though, audition crashed each time (version 3).
-
11. Re: How to get the best vocals out of this poor recording?
therealdobro Dec 31, 2011 11:57 PM (in response to topomorto)topomorto wrote:
topomorto wrote:
You could always split the recording out by frequency band into two files and process them separately.
What do you mean by that?
If you want to do dynamics processing with different settings on different frequency ranges, you either need a multi-band processor, or you need to separate out the frequency ranges into different files.
In Audition, There is a multi-band compressor, but not a multi-band expander.
So, to do expansion/gating, you could take two copies of the file playing simultaneously in multitrack mode, delete or filter the high frequencies from one and the low frequencies from the other, and then process them differently.
And one more thing. What do you think, is it reasonable to use Dynamic processing to amplify the lower frequencies without touching the higher ones, to make the whole track more balanced and also louder?
You could try using the Multiband Compressor to do that job. (that is a multi-band processor, so no need to split up your file in the way I was describing).
That's what modern pop producers do to tracks to maximise the loudness -
A multiband, okay. But why not a lower shelf? Cuz everybody's shy about bringing the bottom end up?
-
12. Re: How to get the best vocals out of this poor recording?
ryclark Jan 1, 2012 9:25 AM (in response to therealdobro)I couldn't use amplification in spectral view though, audition crashed each time (version 3).
If you are not running the AA 3.01 update that should sort out the problem.
-
13. Re: How to get the best vocals out of this poor recording?
Srivas 108 Jan 1, 2012 11:32 PM (in response to ryclark)Thanks!
That solved the issue.






