You are here

Podcast voiceover with lots of ss'ssss

For everything after the recording stage: hardware/software and how you use it.

Podcast voiceover with lots of ss'ssss

Postby studiowolff » Thu Feb 08, 2018 9:16 pm

Hi all,
I’m working on a podcast project. I’ve recorded the vocal in a nice, quiet recording studio with my Auio Technica AT4033 mic + plopfilter.
Even with a deesser i still here a lot of ssss’s
What ever I try, i won’t get it right. Or it’s getting to dull. Boosted every freqentie to check and made a cutting on that spot…but still can’t manage it.
Who can be my teacher, please.
You can hear it at: https://soundcloud.com/studiowolff/love-quest-oxytocin-sample/s-snTz7

Regards,
Jeroen
studiowolff
Posts: 3
Joined: Thu Feb 08, 2018 9:13 pm

Re: Podcast voiceover with lots of ss'ssss

Postby blinddrew » Thu Feb 08, 2018 9:33 pm

I have about 5 different de-esser plug-ins and i've never been happy with the results of any of them* so inevitably I end up going through the track and automating down the 's' sounds. It seems like it would take longer but not when you compare it to the amount of faffing I end up doing with the de-essers and still not being happy at the end of it.


* This is almost certainly a user problem not a software one... ;)
User avatar
blinddrew
Frequent Poster (Level2)
Posts: 3957
Joined: Sat Jul 04, 2015 11:00 pm
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: Podcast voiceover with lots of ss'ssss

Postby Mike Senior » Thu Feb 08, 2018 10:51 pm

Try Toneboosters Sibalance, using its side-chain EQ to focus on the worst-sounding frequencies, and then automate its threshold per instance if necessary.
User avatar
Mike Senior
Frequent Poster
Posts: 993
Joined: Thu Aug 07, 2003 11:00 pm
Location: Cambridge, UK
Mixing Secrets for the Small Studio | Recording Secrets for the Small Studio

Re: Podcast voiceover with lots of ss'ssss

Postby studiowolff » Fri Feb 09, 2018 8:30 am

Mike Senior wrote:Try Toneboosters Sibalance, using its side-chain EQ to focus on the worst-sounding frequencies, and then automate its threshold per instance if necessary.

Thanks Mike!, Iwill give it a try!
studiowolff
Posts: 3
Joined: Thu Feb 08, 2018 9:13 pm

Re: Podcast voiceover with lots of ss'ssss

Postby studiowolff » Fri Feb 09, 2018 8:36 am

blinddrew wrote:It seems like it would take longer but not when you compare it to the amount of faffing I end up doing with the de-essers and still not being happy at the end of it.
* This is almost certainly a user problem not a software one... ;)

You're probably right...i also will give this also try. Just automate the volume on those parts? Like in this article? https://www.soundonsound.com/techniques/techniques-vocal-de-essing
studiowolff
Posts: 3
Joined: Thu Feb 08, 2018 9:13 pm

Re: Podcast voiceover with lots of ss'ssss

Postby Hugh Robjohns » Fri Feb 09, 2018 10:09 am

It's always better to fix the problem at source rather than to faff about throwing ever-more complicated signal processing at it. You can hear sibilance when you're recording, so do something about it there and then!

Sibilance is almost always due to inappropriate mic placement (bearing in mind that some people do have naturally sibilant voices). The mic was probably much too close and directly in front of (or below) the mouth.

So, try moving the microphone up, away from directly in front of the mouth/nose. Move it up to around forehead height and move it slightly further away -- start at a foot or so away, looking down towards the mouth.

Sibilance can also be emphasised by a mic with an extended and/or hyped high-end response, so don't use a peaky capacitor mic; use a dark-sounding ribbon or switch to a ribbon or a moving-coil dynamic.

...And go easy on any compression because that often makes sibilance worse too.

H
User avatar
Hugh Robjohns
Moderator
Posts: 20856
Joined: Thu Jul 24, 2003 11:00 pm
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Podcast voiceover with lots of ss'ssss

Postby Tim Gillett » Fri Feb 09, 2018 11:04 am


Is this the unprocessed file? It sounds as if it might already have been de essed.
Just the unprocessed file would help as more can be judged from it.

People confuse subjectively excessive sibilance from a voice, with sibilance distortion, which often sounds like a lower frequency splattering, and is an equipment distortion issue.

I hear what sounds like some sibilance distortion at first. If that's in the original file, no amount of standard de essing will fix it.

The sibilance distortion is often of a lower frequency than the wanted sibilance itself. The interesting thing about human sibilants is that when we say them, we stop making any other sounds. The letter Z is the only exception. So sometimes carefully applying a high pass filter to the distorted sibilants removes much of the distortion without removing anything else. I tried this on two of the sibilants in your file:

"it's produced when you feel safe."

with, to my ears, much improvement.

I used a HPF at about 5.7kHz . Q = 1.75. You might like to try this as a starting point.

Cheers Tim
Tim Gillett
Frequent Poster
Posts: 1545
Joined: Wed Jan 30, 2013 12:00 am
Location: Perth, Western Australia

Re: Podcast voiceover with lots of ss'ssss

Postby blinddrew » Sat Feb 10, 2018 2:49 pm

studiowolff wrote:
blinddrew wrote:It seems like it would take longer but not when you compare it to the amount of faffing I end up doing with the de-essers and still not being happy at the end of it.
* This is almost certainly a user problem not a software one... ;)

You're probably right...i also will give this also try. Just automate the volume on those parts? Like in this article? https://www.soundonsound.com/techniques/techniques-vocal-de-essing
Sorry I missed your reply earlier, but yes, just like that.
User avatar
blinddrew
Frequent Poster (Level2)
Posts: 3957
Joined: Sat Jul 04, 2015 11:00 pm
Location: York
Ignore the post count, I have no idea what I'm doing...

Re: Podcast voiceover with lots of ss'ssss

Postby Dynamic Mike » Mon Feb 12, 2018 12:12 am

Duplicate the track. Cut out & mute the ess's in the main track, and mute everything apart from the ess's in the duplicate. Then just balance the volume until it sounds normal. It's quicker than automating & you can also be fairly brutal with a de-esser on the duplicate without affecting speech on the main track. (Or should that be 'effecting' the speech?) ;)
Dynamic Mike
Frequent Poster (Level2)
Posts: 2775
Joined: Sun Dec 31, 2006 12:00 am

Non-partisan suggestion for Stroke City: Donderry 



Who is online

Users browsing this forum: No registered users