Harsh vocals

[stefanz40219435]

Hi!  I recently bought a a song in CD quality from Qobuz but I noticed that the vocals in parts of it were harsh and peaked (like you know for example if there is a prolonged "ooooo" or "aaaaaaaa" in the song(when they prolong the end of words) sometimes it sounds harsh and unnatural. I can't explain it better sorry. Do you have any idea how can I fix this in Adobe Audition if I can at all. It's probably bad recording or something but if you got tips on this it would be cool. I tried compression but the results aren't very satisfying,the whole song gets quieter. Do you know what effect could fix that and make the voice smoother but in the same time alter as little as possible the music and the whole song itself.

[SteveG_AudioMasters_]

It's very unlikely that a commercially recorded work would sound that bad, in my experience - although not absolutely impossible I admit. But without actually hearing the work in question, there's no way we could give you any information about rectifying any aspect of it, simply because all audio is different; no one-size-fits-all solution for anything like this. Also, what you monitor on can make a significant difference to your perception, and it's very easy to alter something to compensate for deficiencies in that, and have it sound lousy on everything else you play it on. So at the very least we'd need a reference for the track in question...

[stefanz40219435]

Hey so I completely forgot about this question,I found out since then that listening to that song it sounds way better if I put the 8k on the equalizer basically at the bottom then the problem's fixed but evrything becomes more muffly naturally. So if you are still on that here's the track. Listen to 52 seconds the oooo there. and there is the fix I found. It's Bulgarian but I think you will hear it it's nothing to do with the language it's basically let it go in bulagrian.

[SteveG_AudioMasters_]

I downloaded the track and yes, it sounds muffled, and no, it's not a brilliant vocal recording - well I don't think it is anyway. But... it's not that bad. So I put your 12dB at 8k back, but rather more importantly normalized the track back to -1dB because it was clipping quite badly - for all I know that could easily have accounted for your harsh sound. Anyway, on some very fussy Neumann monitors in an acoustically controlled environment, I get no harshness on 'ooo' sounds at all. And believe me, if there was, I'd hear it...

[SteveG_AudioMasters_]

Whilst it was sitting on the 'other' machine, I did a little experiment to see if the vocal could be improved slightly, because the one thing I did notice was that the vocal was almost sibilant in places. So I knocked that out where it happened, and also increased the vocal level very slightly in the track, as it sounded too much like a struggle the way it was. Oh, and the stereo was a bit narrow too, so I widened that slightly and added a bit of warmth to the whole thing. To me, it sounds better here, although YMMV...

[stefanz40219435]

Wow it's amazing it sounds much better than the original much cleaner on the belting and the voice is somehow more powerful and upfront and you can hear the details in it more which is great.Thanks! I will prefer to listen to your edit honestly. The only thing is maybe it's the wider stereo but when there is like a drop when everything hits hard for example in 3 minutes 1 second it sounds tinnier for some reason thre isn't that depth that base to it for some reason I don't know,otherwise I like the wider sound stage and the vocal is way better here honestly I love it. I think you did better than then sound enginners or whoever made that production for Disney Bulgaria which is crezy since Disney has prefect production everywhere else in the world but for some reason here sometimes it's actually pretty bad,they should change their team maybe make some changes to it. I guess I could try adding base with EQ to make it more basy on the drop but could you try changing nothing on the edited version that you sent me but narrowing the stereo again so it's as narrow as the original to see if that will be better because in evry other aspect it's perfect!

[stefanz40219435]

also I am saying that of the stereo also because sometimes the vocal is in both ears but out of sync for some reason and it's a abit weird so maybe I like narrower stereo I don't know otherwise everything else I love

[SteveG_AudioMasters_]

The stereo width is about right - both as measured and listened to. The reason that the vocal sounds strange when it gets louder is because I only did a quick job on it - that was a 'one setting for everything' pass, which was also pretty much what happened to the original. To really make it sound better you have to break up the treatment into sections and treat them appropriately. By the way, most of this you can't do directly with Audition - this was done mostly using iZotope's Ozone 9. I might have a look at the louder bits - I don't think it would take much to improve them a bit. The other thing is that this is a loudspeaker balance, not a headphone one, which may make a difference to the way you percieve it...

[stefanz40219435]

Yeah,true,otherwise the stereo width is definitely wider and sounds better like that because there is a bigger soundstage,just sometimes(that might be my headphones,there is a slight difference between left and right ear so it sounds like it's panning really quickly.) But otherwise you did and amazing job thanks a lot! Now I can listen to the song without havinng to bring down the highs whivh muffles everything. I just added a bit of base in the EQ but that's my preference,because my headphones are flat anyway(Audiotechnica ATH M70x) becuase they are monitors. I will be listening to the edit now. Disney Bulgaraia should really hire you for their productions(they are pretty bad for some reason when it comes to the frozen movies-the second movie's songs in Bulgarian also have problems-like the chorus dips and is quieter than the before chorus parts(the vocals at least) which is bad. Also yeah if we had access to the stems this would have been way easier to edit working on a wav that is already mixed is difficult because you are changing everything you can't just apply something to the vocal and somethig else to the other instruments. Honestly if I had the stems I wouldn't even need help because I can mix in Pro Tools but I don't,so I can't really fix a mixed file very easily.  You did a rwally impressive job considering you didn't have access to individual channels,thanks a lot really!

[stefanz40219435]

if there is a follow up you can send it to me if not thanks a lot! I could also work on the louder bits to see if they can be improved but it honestly sounds almost perfect.

[SteveG_AudioMasters_ · Accepted solution]

Since it was still there, it occured to me that if I swapped the processing order slightly I might be able to optimise it a bit better, especially at the end. Part of the problem is that she moved back just a bit too far from the mic, which altered the vocal perspective in a way that didn't match the backing properly - especially on that long-held note. I split the vocal perspective adjustments into four parts, which now match the orchestration slightly better. I've juggled the stereo perspective slightly differently - hopefully this won't upset you so much - and done a classic 'scoop' on it in two goes, which improves the bottom end, I think.

But you are correct; without the stems there's a limit to what can be done, and I think I may have reached it...

[stefanz40219435]

Wow it basically sounds perfect now I can't ask for more from it! Thanks a lot for dediacting your time for this,I literally couldn't stand to listen this song as it was and couldn't fix it myself. Now that I compared yours to the original the original sounds like garbage compared to the one you did the second one. Thanks a lot for real!

[stefanz40219435]

Don't get me wrong the singer and the song were amazing but the mixing was terrible now it's finally amazing as well!