RMS Normalize seems to undo everything else

Oh dear, I think I just got confused. So correct me if I’m wrong: the file was better and it was only a wee bit way off, as opposed to at the beginning of this post.

Okay and it passes!

So my problem was… how I was recording and taking bad advice off of Youtube. Lesson learned! Now this passes Amazon’s requirements. And the plus is I understand RMS… okay not very well but I get it a bit more. Sort of. Better than nothing.

But I can tell you this file probably wouldn’t pass a voice acting review in some places, so now I have to concentrate once more on the mystery of volume peaks and other things. See I’ve been trying to put together my demo reel well for a long time. Mixing files from different shows and getting it even has turned into quite the challenge. But, that’s for another thread I’m sure.
recording result.jpg

I think I just got confused.

I’ll bet real good money Steve was measuring the wrong sound file.

Your original announcesample.wav is perfectly good for a live reading, but isn’t as perfect as Steve makes it out to be. One quick trip through the three-step mastering suite and you should be good to go. Yes, if you want to get obsessive, you can add additional gentle noise reduction and a few more bells and whistles. But post-mastering passes by itself.

it could be said that the least amount of processing is the best amount given you’ll be doing it for years. It makes my teeth hurt when new users arrive with a laundry list of effects and corrections and they haven’t put voice into the computer yet.

RMS (loudness) and peaks are built into the suite. You don’t have to worry about any of that. It’s a corrections suite (harmonious grouping).

Now if you’re after creating a reel out of many different performances, good luck. I’ll watch. We can’t take processing out of a performance. This is where good speakers or headphones become terrifically important. Past basic recording hygiene, it’s all about the sound.

I need to sit and read your publication if it’s not already obvious I’m missing many of your questions.


Just to bring this around. Steve told the Noise Reduction tool to reduce the background noise by about 6. When I measured it, I got -68dB (third number). After additional processing, the noise is about -75. Subtract those two…

You can’t do that forever. Noise Reduction affects your voice, too, and too much processing will fail ACX.

Please note we don’t have official YouTube training. It’s too hard to do right, it gets left in the dust too easy, and nobody wants to go through the editorial process. So. Almost all of the video postings are wrong, misleading or for a very old version of Audacity. Sometimes all at once.

Also note ACX processing is only the first half. It also has to make it through Human Editorial Inspection. That’s where voice quality and reading artistry becomes important. So, you’re now at the artist presentation starting line.

You can be theatrical, but you do need to back away from the microphone when you yell, etc. etc. etc. I once played two different people (of different sexes) by clever microphone management. Directional microphones sound very different when you’re up close.

Lauren Bacall’s voice was lower than Humphrey Bogart.

“You know how to whistle, don’t you Steve?”


I actually didn’t do noise reduction in the last take. It didn’t need it - and noise reduction doesn’t take out those mouth smacks and tiny sounds you do when the mic is, well, where human ears aren’t normally. So I went through it today and took those out manually for a better file - which is good because listening to it again helped me to catch where I’d repeated myself in reading. Oy vey!

As for putting together my demo reel - well. That’s for another thread I suspect. :slight_smile:

For the first time I actually feel hopeful about the entire getting the first recording down process, so that’s something. Da da da da. You found: sound recording booth. Da da da DA da da… You are now level (1)!

We have a good track record for getting people published.

If the microphone is crisp and harsh, sometimes a sock or hanky over the microphone can help. People make microphones with a high pitch boost because it “sounds more professional.”

Whatever you say.

That boosts every tick, pop and mouth noise.


Well I certainly am grateful for your track record. But what is a high pitch boost? The mic picks up the sounds more in the treble area? Wouldn’t that change how a baritone sounds when recorded?

The mic picks up the sounds more in the treble area?


Wouldn’t that change how a baritone sounds when recorded?


Blue isn’t the most forthcoming about technical specs, but apparently, the Pro model is pretty flat and accurate over the audible range. No serious boosting.
Screen Shot 2018-03-30 at 18.31.18.png
Compare that with the Shure rock band microphone.
Screen Shot 2018-03-30 at 18.34.50.png
Those hills over on the right are the famous “presence boost.” Crisp, hard sound whether you want it or not. Somewhere between 3,000 and 4,000 is infants screaming on a jet to give you a reference point.

So, no, you should be OK.

Do experiment with a sock, though. Experiment. One poster got good, balanced sound with a thinner polyester sock. It’s worth it if you don’t have to correct each and every lip smack. That’s going to get old on your fourth book.


Well I have a pop glove on it. Like this one: https://www.zzounds.com/item--MUP58WS?siid=1082&-D1R8rrImrZvvS6JqiuGoCMICZDr728Z73sQqbV0YkExxoCkiUQAvD_BwE=

I guess I could put a sock under the filter? Or is that what the sock is for? Just making sure.

Well I have a pop glove on it.

Did you do that to solve a specific problem? That design is supposed to help with wind noise and give P pop suppression. It does it by restricting large movements of air. In English, it suppress bass notes.

I’m not a fan of doing that unless you’re solving a problem. What’s the sound difference with and without? Do you have that black tennis racket thing, too?

You do have your Sony MDR-7506 headphones, right? You’re flying blind (deaf) without good monitoring. I do not have a pair of those. I don’t particularly like them, but Hollywood adores them and they will tell you when you’re doing something wrong.

That’s them on David Greene’s head in NPR West.




Well, it would be nice if it stopped the plosives but no. No it does not. It’s just there to be pretty. It’s blue. My office is very colorful.

The tennis racket thing is there also, but that thing I hate. It never stays where I want it to.

Mic technique has been my answer in this. I have the mic to the side from my lips. It still records well, especially in mono, and there are less plosives that way. The smacks and wet sounds remain though.

My headphones are cheap logitech ones. Those other headphones look sweet, though!

I have the mic to the side from my lips.

That works for me, too. Off the corner of my mouth instead of straight in front of my nose. If I do that just right, I don’t need the tennis racket.

P Popping and plosives work by sudden broad masses of air. The black tennis racket works by only stretching enough to pass the higher sounds but not the lower ones. That thing David Greene is using is different. That one is a stiff grid in front of the mic. The holes in the grid aren’t big enough for P Popping.

Did you try the socks yet? Remember this is a forum. Users helping each other.


Since I’m sure you have loads of leisure time, do a clean recording (without the blue foam thing) and then two different socks, announcing as you go.

“Here’s heaving skiing socks, here’s casual wear.”


LOL. Oh yeah, loads. But it’s loads of laundry, not time.

I wasn’t sure just how to do the sock method so I decided to look it up first - and found this neat video. https://www.youtube.com/watch?v=GN03wwXiS3k

I realized… maybe… what I’m thinking are plosives aren’t plosives. Maybe I’m just really noticing that p sound. Paranoid of the popping.

I know it’s not a help desk - the reason why I’m usually so silent here is time. I was working anywhere from 12 to 16 hours a day for a while. I’m only just now signing off from clients and backing off from being a “good citizen” so I can make time to learn how to record better. You guys have always been great here. It’s one of the first places I haunt when looking for answers, even if I rarely talk.

The past couple of days have been me trying to put together a new character demo reel, now that I was armed with your glorious information. And guess what. I got things reasonably level, although I’m still unsure about peaks.

and found this neat video.

Yes, but.

This isn’t a Pop question. We know how to do that. Did you notice his overly crissssp ssssibilance went away with the sock? So that’s the answer.

Did he ever tell you what kind of sock he used? Inquiring minds want to know.

We’re not fond of YouTube videos. You’ve had really good luck, right?


I’m still unsure about peaks.

Theatrical peaks or blue soundwave peaks?

ACX has firm rules about that because they have to be able to transform your work into other platforms and formats with no hint of damage. Conversion in and out of MP3 can cause peak damage. That’s why Effect > Limiter in the ACX Mastering Suite is set to -3.5 rather than -3. And that’s also why you need to pay attention when you apply the tools.

A note here. Inside Audacity, peak rules don’t apply. Audacity uses a weird data format that doesn’t overload. That lets you do whatever you want without worrying about it. Outside of Audacity, the world comes to an end at 0. Once your sound goes up to 100% or 0dB, it’s going to be audibly distorted and there’s no way to recover. That means the microphone before Audacity has to be cool and your client export after Audacity has to be cool.

As a fuzzy rule, we recommend no peaks over -1dB for a client export. You can run Effect > Limiter (by itself this time) reset to -1 on any performance to make sure it doesn’t go over. That should be one of the last things you do to the show.
Screen Shot 2018-03-31 at 14.53.05.png
Did I hit it?

Have you figured out which socks you’re going to use?


Okay, I have to level with you here. I might go to -6db for this. But. M’kay, so.

I don’t wear socks.

As a matter of fact, I will go days in a row without wearing shoes. Not even my fox slippers the other half got me to encourage me to, well, wear shoes.

There are FANTASTIC perks to working at home, and to be brutally honest: not wearing those leather shackles of modern mundanity except once in a while is one of the best of them.

I was fully prepared to steal my husband’s … icky… socks for the experiment. I really was. If I could find a pair he hadn’t been wearing for days in a row, but then I began to question if I were to slip the sock over the mic like a … sock or perhaps I would wrap it around and secure it in order to get the maximum sock thickness in place. That’s why I found the youtube video. I was too embarrassed to ask. :-p

Usually I actually do get lucky with youtube videos, but then again I am known to search carefully for a long long time on odd days.

That being said I’ll steal a couple of socks when they come out of the dryer and give it a try. When I DID have socks the other half stuck his over sized feet into them and ruined them, so let’s see how he likes the shoe on the other foot. These will be black no-brand-name socks.

No I didn’t notice where he was aiming the puffs - it looked to me like it was dead on. As for his crispness, I noticed right away his voice seemed to get fuzzier and fuzzier the more fuzziness he added. I thought to myself that I wouldn’t like that at all.

Peaks: What I was told was “peaking,static,Your audio is still way too loud. Try bringing it down a bit. (denied)”

Now I understand a little of what was still too loud, although they didn’t say it was loud to begin with. This was not my first submission.
The static wasn’t static. It was a seltzer sound effect in the scene I had chosen to use for the demo.
Peaks were also a new one on me, as the other times no one had complained of peaks. Looking at demos that passed, well… they have the same “shape” with what could be peaks. So I’ve been going in circles trying to understand that ever since.

Just random experimentation at this point: No one uses leveler anymore according to this forum. But all of the voice over information says the same thing: even out your file sound.

I’ve also been reading on the benefits of using a noise gate over noise reduction. I tried it and really do like how it cleaned things up without messing with the overall tone.

So I added two steps: 1. noisegate set to work at the very bottom where I saw the needle moving in the quiet parts. 2. Leveller, set on the heaviest setting under manage. It was the only one that actually leveled things out, as it were. Then I went through the rest of the motions.

The result is I have a file that sounds, well, more level. I’ve got my headphones on and I think it sounds good. Very importantly: I hope it doesn’t have any sound peaks. And it passes ACX.

For the record, I did try using the compressor for it but I am so bad at this that leveler was easier for me to understand. LOL

Whatever works for you.

Yes, there are some serious problems if your voice wanders in volume over the course of a chapter. The standard suite can withstand a little of that, but any “heavy theatrical wandering” is going to be a problem. This is one reason you’re wearing good quality headphones during the performance. It tends to even out your voice as you read. And the admonishment widely ignored, you’re supposed to keep and eye on the Audacity sound meters like the recording engineer would be doing.

I’m sure you noticed Leveler is still included in the standard Audacity package. It’s under Effect > Distortion. It was going to be left by the side of the road bleeding in a past version update, but we rose up as one to say it’s hands-down the best way to made Taxicab, 2-Way Radio and Airline Pilot voices. So it was retained as a special effect.

It was a seltzer sound effect in the scene I had chosen to use for the demo.

They are fond of saying this isn’t radio theater. We’re not brought to you this week by Chesterfield Cigarettes. Dum Dum Dummmmmm. It’s a plain reading. I think the most exciting thing you’re supposed to do is voices so the client can follow the characters.

So that’s why your submission failed. It may have had volume and peaking problems as well. Did it pass ACX Check, or was this before you got your hand in. Once Human Quality Control starts seriously analyzing, they’re going to find all kinds of odd things wrong.

Theater is what you do for your podcast. You are going to do a weekly podcast, right? You switch your microphone to bi-directional and put two actors on each side.
Screen Shot 2018-03-31 at 20.30.38.png
Simon Whistler does a podcast I like. Turns out he does three podcasts—that I know of. I counted the production requirements and it’s around 30 hours…a day. [Puzzled Spaniel Look]

Oh, the sock. That’s the metaphor. Do you have Railroad Engineer Bandannas? That works. How about Ikea “Tekla” dish towels? That would work. The tradeoff of slightly muffled voice is not having to constantly go through and stomp out sibilants and lip clicking.

There was a phrase here… “Wet Mouth Noises.”