Recording levels

Special Interest Groups Audiobook Production
1

Hi everyone, I’m new to Audacity and I’ve been fiddling with the acx plug in that allows you to check if it passes. Without fail, I always seem to be just missing the RMS check. I’m always anywhere from 26-32. Is there any way to just set all the levels I need before I start recording so it just meets all the specifications I need? When I turn up the gain I pass the RMS but fail the noise floor check. I’m using a Blue Yeti Pro, if that helps with info and I use a macbook with the latest operating system.

If anyone is so kindly willing to walk me through the process of setting all the acx levels I need, I would be so eternally grateful, or pointing me in the right direction. I’ve read so many other posts but fear they are above my head at one point or another.

Thank you so much, I really appreciate any and all feedback and help.

:slight_smile:

2

The short answer is no, and your experience is perfectly normal.

We recommend peaks at about -6dB on the bouncing light sound meters and 50% on the blue waves as you record (they measure the same thing different ways). That’s so you have room to screw up by getting too loud or too soft. Too loud in particular is a problem. If you go over 100% or 0dB, with your voice, that will create distorted crunchy sounds and they’re permanent. You lose.

So right away, that makes you different from ACX standards which has sound peaks bumping around -3dB or so.

You have one strike against you immediately. Very few people pass ACX cleanly and easily with a USB microphone. In order to smash microphone, preamp, mixer, power supplies and USB digital conversion into one convenient, inexpensive package, they had to take shortcuts. One of the shortcuts is low volume. You may find that it’s really difficult to overload the microphone with your voice. That’s by design. From above, overload produces permanent distortion, so, much better to go low volume and fix it later.

It’s the “Fix it later” that gets to be a problem. As you’re finding, high background noise and low voice volume are Not Good. You can generally meet two of the ACX specifications, but not all three.

As an experiment, I did record a voice test and got it to pass ACX by adjusting the volume. I kept telling people you could do that, so I sat down and did it—but I wasn’t using a USB microphone. The illustration is a rock band microphone, small analog sound mixer, and digitizer. On my Mac, I don’t need the digitizer. Oh, and a quiet room. That’s a big deal. You can’t do this in the kitchen or any noisy room with echoes.

So.

Post a sample of your work, and we’ll take a stab at corrections. Sometimes, we can make suggestions as to how to get closer without patching, filters and effects.

http://www.kozco.com/tech/audacity/TestClip/Record_A_Clip.html

That silent patch is important. Do Not Move.

Koz

3

Do not adjust the work. Finish talking, Stop and Export.

Koz

4

Hi, thank you so much for all that information- it made sense to me!! My husband has been playing with a recording for hours tonight and finally got a small snippet to pass ACX… ugh, hours lol! Our yeti pro came with an analog attachment cord. Based on what you said about them craming that all into a prepackaged mic, do you reccommend getting the mixer? If yes, can you reccommend an affordable one or the name of that one pictured above?

We are recording in a large walk in closet with a door, high pile carpet, bookshelf and clothes… we hoped the more stuff the more it would dampen any sound. Though, we turn off every possible thing in the house like heat while we’re in there.

I will post roomtone and a recording tomorrow, asap!

Thank you so much, your insight is priceless in this journey!!

5

finally got a small snippet to pass ACX

Probably not. ACX also has a failure they call “overprocessing.”

When you submit to ACX AudioBook, it goes first to what I call The Robot which does about the same thing our testing program does. If you pass basic competency, it then goes to actual human Quality Control. That’s where it dies if you spent hours beating your submission with a stick.

There are tricks to this. Give us a shot at it before you go nuts. Use the regular USB connection. Because you have a Mac, you’re probably not going to have The Yeti Curse. I’ll tell you about that later.

Making a studio is a good start. I got insanely lucky. The first owner of this house had a son that played drums, so the third bedroom is soundproofed.

Koz

6

Hi, here is a short clip… I hope you’re able to give me some insight, tips, and what we may be doing wrong and needs improvement!

Thank you :slight_smile:

PS
Is mp3 the correct format to just share it on here?

7

Is mp3 the correct format to just share it on here?

No. It’s not. We’re not just sharing. We’re going to inspect your work for correct, technical accuracy and MP3 adds compression sound distortion. So WAV (Microsoft) it is.

If you closed Audacity without Saving an Audacity Project or Exporting WAV (Microsoft), then you need to read the test again. That’s newbie mistakes 2 and 3.

Welcome to reading for AudioBooks!!

You would think all this would be written down, wouldn’t you? You can’t. Everybody starts the process with everything different. Generic instructions get so vanilla and non-specific, they stop helping.

Pay for studio recording time > Submit audiobook sound tracks > Bask in publishing success.

Koz

8

I got your existing clip to pass by applying several different tools, but gently. I can’t tell what really happened because of the MP3.

Koz

9

Ah, that’s interesting! Newbie mistake, all my files are mp3 so I will have to rerecord and save as wav, yay!! Well, that’s good to know!!

Is there hope or do most people buy studio time?

Thanks for taking a look/listen at my clip, I’ll post an appropriate file as soon as I can.

10

MP3 is an end-user sound format. Make an MP3 for your Personal Music Player and Full Stop. There is no ‘open up an MP3 and do something else with it’. The quality can seriously turn to trash when you do that. WAV (Microsoft) is a good format for production and archive. My pro paid recordings were all done in WAV.

Yes, I know ACX requires submission in MP3, but if you read their submission standards, they have a lower quality standard but not upper. You should submit in very high quality MP3—made from your super, perfect quality WAVs.

Is there hope or do most people buy studio time?

There’s no good way to know because they tend to not need the help forum. We did get one poster who was going great guns with her studio career cranking out work after work…and then decided to do it at home. How hard could it be?

Nice voice, by the way. This should go well when we get the system to record you right. You don’t seem to have background noise or excessive hiss or any of the other problems of a poor room, environment or microphone.

Your voice is very slightly “thin” which is what’s probably causing the loudness to fail. I did “first pass” volume setting, noise reduction, and compression. I’ll post the process when I get the real one. That sounds terrible, but you get used to the swing of the corrections. It’s a little like the four volume set “how to ride a bicycle.” The first time you hit it…

This is the English description of the three ACX numbers.

https://forum.audacityteam.org/t/measure-between-23db-and-18db-rms/32770/16

Koz

11

Cool! This is all so extremely helpful. I just spent an hour reading some of your other posts and helping out other people and learned a ton. I also read a blog post by someone else about building a foam box around your microphone to assist with being able to turn up the gain without crashing the noise floor. Do you recommend this for someone with a “thinnish” voice such as mine? Could that help pick up my voice better? I fear if I speak any louder I’ll lose the authenticity of my natural speaking voice.

It is a relief to know there is nothing hissing or daunting in the background!

It has been raining cats and dogs and my recording closet is on the 2nd floor… waiting for it to die down to re-record, bummer!

Thank you!!

12

In My Opinion you should stick with what you have and see how well we can do with post production filtering. Like I said, you have a perfectly good presentation and I don’t hear anything wrong behind you. Many people would kill to be able to say that.

A note on that forum test clip. It’s a Really Big Deal that you don’t make any human noises in the first one or two seconds. The instructions say hold your breath. After that, you can gasp and do whatever you want for 18 seconds. If it’s more comfortable for you, put the Room Tone/hold your breath at the end.

One poster insisted on shifting in his seat and scratching himself. We could hear everything. This is the tiny sample of what the microphone and room are doing with no sound from you. You have to be there because your body affects the room sound quality.

The forum will reject anything over 2MB. That works out about 20 Seconds of mono WAV.

Koz

13

Hope you don’t mind me joining in the thread. I too started out only a month or so ago to record my two books for Audible submission.
This is the equipment I use.
Mac mini 6.2 - 2012, Intel Core i7,2.3 GHz, Memory:16 GB. El-Capitan OSX 10.11.3 2TB
Audacity 2.1.1 - Focusrite SOLO Mini Preamp - Shure55SH Series II (Dynamic)

I have a few posts scattered around this forum, so I won’t repeat it all here.
But without getting too technical, I discovered that if I get the NoiseFloor down to as low as I can - here it’s between -70 and -60 usually, and I’m only doing this at home too, then I record with my mouth about two or three inches - it varies of course, from the mic.
Then I edit the recording - getting out long breaks, and any other “bad bits” like mouth noises, then I check it with ACX check.
Mostly I then only have to use Normalize at low levels, usually say -3db to get the RMS right. If I have to Amplify to get the RMS up to where it should be, I then use Limiter to set the peaks to -3db
So like Koz, what I’m saying is minimum processing. Really, all I ever use now is Normalise, and if I need to pull the peaks back, Limiter. Then export as a WAV file. Always save your project.

I haven’t ever considered getting a professional to do the audio for me - If I could afford that, I wouldn’t be doing it in the first place. Plus, I enjoy the challenge of getting it right.

14

Hi, rach thank you so much for your suggestions and tips- I am going to try them out tonight!! I have been trying relentlessly to upload my room tone and audio clip in the appropriate format but it won’t upload to this site. I live in the middle of the country and I’m thinking my internet may be the issue. I’ll be heading out of state this week so hopefully while I’m at a hotel I can get fast internet that will let me upload for more suggestions!

Thanks so much to you and Koz!

15

Do you get an error message? A 20 second stereo (two blue waves) WAV clip will not fit on the forum. It’s 10 seconds stereo or 20 seconds mono. Most voice work is mono. If you’re part of a theatrical presentation with effects, environment, etc, then yes, you need to be in stereo. But not audiobooks.

The forum has restrictions on file types as well. If you’re not actually in WAV (Microsoft) or MP3, it may not like that.

There’s a way to force Windows to show you the filename extension. Look for something.wav. or something.mp3. Try for the wav version. That file type sound tells us more.


Koz

16

Hi LadyLee04, glad I could help.

As for uploading, it’s very limited in what you can upload. WAV files are gigantic, because they pack all the recorded info into them. But that’s what’s needed. However, to get a sample uploaded, you need to Highlight the track you want.
Export Audio will be greyed out unless you select the track to export.
Make sure it’s
Mono
max 10 seconds

Highlight the track, and export to WAV, PCM (Windows) 16 bit PCM

For example: this first track is 4 seconds long. It’s mono wav.
The second track is 7 seconds long, mono, wav.
First is 361k in size
Second is 615k in size.

Below your post, you will see [Options ][Upload Attachments]. Select Upload Attachments.
Choose a file - This is where you click on that and it will open explorer/finder so you can select the file you want to upload. When you find it, select it.
When it’s name appears in the little box, then click Add the file, and wait for it to upload. It will blink back after a few seconds and show you the attachment below your post.
I’m not sure but the limit may be two files. However 1 will do for Koz to look at.
When that’s in place, just [Submit] your post.

You can see the attachments and sizes etc below here.
hope that helps …
Robert

17

Ah! That rings a bell, I believe I was supposed to do that and then made a too long clip- I will re do it again lol. I’ll get it right!!!