Humble Request for Workflow Advice

Narrating and Producing Audiobooks.

ImageIf you require help using Audacity, please post on the forum board relevant to your operating system:
Windows
Mac OS X
GNU/Linux and Unix-like

Humble Request for Workflow Advice

Permanent link to this post Posted by dottedllama » Sun Mar 12, 2017 12:36 pm

Hi all, I’m looking for your professional opinions on my mastering chain so I can make sure I’m producing for ACX to the best of my ability. I’ve been lurking on these boards for ages and regularly trawling through YouTube and Facebook groups to learn the basics and I think I’ve got an ok grasp and a pretty ok product now, but am not sure if my mastering chain is as efficient (or giving the best sound) as I’d like it to be.

Firstly, the basics. I’m operating in an sound insulated DIY booth and my general noise floor is around -80db. My windows 10 laptop sits outside the booth and cables for my monitor, keyboard, mouse and pre-amp run through a tiny (and insulated) hole through the bottom of one wall to reduce the chance for any fan noise. My recording chain looks like this; Samson CO1 studio condenser into a Focusrite Scarlet Solo pre-amp, then into the laptop USB. I record using both a pop filter and a mudder on the mic itself because I can have issues with plosives (that I’m working on through proper technique!). I’m using Audacity 2.1.2.

Besides plosives, I also struggle with the dreaded mouth clicks which I have managed to reduce and still have some technique work to do there. To give you an idea of my raw sound, I have attached a wav clip with a few seconds of room tone at the beginning. You can hear the mouth clicks etc which I will normally correct either through a combination of spectral editor and the “draw” tool, or if it is in a spot I can’t quite get to, I leave it for my mastering chain to resolve.

My mastering chain flows as such:

1. EQ “LF Rolloff for Speech” (this has the added bonus of reducing the sound of plosives in my voice)
2. Paul Licamelli’s De-clicker with a few minor adjustments on the settings
3. Compressor (2:1 w/threshold at -30db & -NF at -40db)
4. Normalise (to -3.1db and remove dc offset ticked)

When finished, this normally raises my noise floor to around -62db so I use the noise reduction tool with the 6/6/6 setting to try and get it down to -75 or less. Sometimes I need to do an additional pass with the noise reduction using 3/6/6 before it gets there. Then I run ACX check to confirm it’s ok, and export to MP3. I’ve attached a wav of what the same file sounds like after running my mastering chain and noise reduction.

So my very humble request is to get your opinion on whether this workflow is best for my voice / set up and whether there is anything I can do to further improve my sound for future projects?

Thank you in advance guys, I am beyond grateful for this forum!

Clip samples:
https://soundcloud.com/tj-richards-9346 ... ip-samples
dottedllama
 
Posts: 2
Joined: Fri Oct 21, 2016 3:29 pm
Operating System: Windows 10

Re: Humble Request for Workflow Advice

Permanent link to this post Posted by kozikowski » Sun Mar 12, 2017 2:53 pm

I don't understand the two clips. The second one is quieter and I favor the first.

It's a pleasant voice with good theatrical presentation. It sounds like other audiobooks I've listened to. If it passes ACX-Check (and I don't see any reason why not), it should be good to go.

A note that absolute perfection is not necessarily a desirable goal, particularly if you have to get there with a lot of work in post production. Everything you do and each chapter has to be patched with each step and that can get old as your career takes off.

For example, did you use the 3dB Noise Reduction step because it sounded like you needed it, or it gave you a warm feeling to get the noise analysis value lower? The ACX limit is -60 and yes, I understand some noises can still be objectionable at that value, gentle microphone hiss is sometimes OK. Also, it's possible to bump up the first noise reduction to 9/6/6 and eliminate the second pass. 12/6/6 is the limit. Beyond that, your voice starts sounding brittle with correction artifacts.

Also, once you settle on a patching routine, you'll have to stick with it in order to satisfy ACX's requirement that all your chapters match. Changing your correction suite chapter by chapter may not be a good idea.

Koz
kozikowski
Forum Staff
 
Posts: 36710
Joined: Thu Aug 02, 2007 5:57 pm
Location: Los Angeles
Operating System: OS X 10.9 Mavericks

Re: Humble Request for Workflow Advice

Permanent link to this post Posted by dottedllama » Sun Mar 12, 2017 3:20 pm

Haha I will admit to a certain amount of warm fuzzies when team production values are lower. I did the second 3db pass because I thought having all my noise floors between -75 and -78db would be more likely to pass through the ACX system quickly. I think having studio monitor headphones makes me super picky about sound. Perfection is a hard mistress!

Thank you for taking the time to listen and respond it's much appreciated.
dottedllama
 
Posts: 2
Joined: Fri Oct 21, 2016 3:29 pm
Operating System: Windows 10

Re: Humble Request for Workflow Advice

Permanent link to this post Posted by kozikowski » Sun Mar 12, 2017 5:40 pm

If you are one of the angelically bless people who can record an audiobook track with little correction, your noise floor will generally settle in the -63 to -68 range. That's with microphones and interfaces that normal humans can afford and, like you, with a good recording studio. Any track with a -60 noise floor or lower will go through ACX technical acceptance.

You should know that ACX has not certified our tools. We're going by the specifications they published and the forum's success rate. You should also know that ACX Check is only the first step. They are also going to subject you to theatrical quality—Human Quality Control. This is where you fall apart if you only passed technical testing by beating your voice with a stick. If you submit something that sounds like a poor cellphone in the blackness of space, they will probably reject it. They have a failure called "Overprocessing."

That's why I was clear your voice sounds natural and theatrically enjoyable.

And just when you dust off your hands and call it a success, we had one poster rejected because she submitted the wrong number of seconds in her introduction format.

You do have to pay attention. I assume you've read through the ACX submission instructions?

http://www.acx.com/help/acx-audio-submi ... /201456300
http://www.acx.com/help/setup/202008260
https://www.acx.com/help/encoding/202008310

These are samples. you should go back to the top of their web page and make sure you don't miss anything.

Koz
kozikowski
Forum Staff
 
Posts: 36710
Joined: Thu Aug 02, 2007 5:57 pm
Location: Los Angeles
Operating System: OS X 10.9 Mavericks


Return to Audiobook Production



Who is online

Users browsing this forum: No registered users and 2 guests