Audiobook Narration - Setup & Waveforms - Send Help


Complete DIY beginner here which is about to be so so obvious… I’ve done some VO work for commercials in a studio, but was never required to edit/master my own files. I’ve watched teams edit files and worked alongside them, but now, I’d like to get started with some narration and VO projects from home (if I can manage).

I was gifted a Rode NT1-A starter kit after a project wrap by the studio I worked with and have been playing around with it for several weeks now. My home recording space is still a WIP.

My current setup:

  • Focusrite Solo 3rd Gen / Apogee One (currently borrowed)
  • Rode NT1-A
  • macOS Monterey 12.0.1
  • Audacity 3.1.3 / Garageband 10.4.6
  • Various sound shields / insulation cubes

Current challenge(s):
My waveforms are extremely small. When I speak into the mic to test levels or record, I am consistently stuck at -24db to -18db. I rarely make it to the recommended -18db to -6db range for ACX (for example).

I have tried tweaking the gain, maxing out the gain, adjusting the inputs in Audacity, using the ACX mastering macro from the forums. When I turn the gain to max on my solo, the sound quality is terrible and my voice is garbled. I watch for the lights etc on my interface but still struggle to get consistent higher volume input. The Audacity macro for ACX mastering is great though.

I have also tried to adjust the mic inputs from my system preferences on my Mac, but don’t seem to have the option.

System settings:
Screen Shot 2022-03-23 at 8.38.27 AM.png
Screen Shot 2022-03-23 at 8.38.39 AM.png
Audacity settings preview:
Screen Shot 2022-03-23 at 8.56.26 AM.jpg
The larger wave forms in the first test are after ACX mastering macro. The others are not edited/mastered.

Similar problems when I use my Focusrite Solo interface. I can’t seem to get the input into that ideal range. I am not particularly soft spoken, but perhaps I am for it to be an issue.

I have also tried Garageband with similar results. I have used presets shared by Rob Drinks, but still have challenges with the waveforms and input levels.

Sample file:

Am I the problem? Considering two different interfaces produce the same results. Perhaps this mic is not for me? Any suggestions or recommendations from soft-ish spoken narrators? Am I complicating things too much? Would a simpler set up better - i.e: a USB mic like the Blue Yeti or the Rode NT-USB Versatile Studio-Quality USB Cardioid Condenser Microphone?

Any and all help welcome! Thanks so much in advance.

If this is happening with both interfaces I suspect the mic or the mic cable.

So if you get up-close to the mic and speak loudly or shout, can you ever get the LEDs on the interface to show clipping?

I assume you’ve got phantom power turned-on, but does the Rode have an LED to show it’s actually getting phantom power?

Low levels wouldn’t be too unusual with a dynamic microphone but a condenser mic usually has plenty of output if you’re speaking (or singing) close to the mic with a strong voice. (Many interfaces don’t have enough gain for a dynamic mic.)

And since you do have experience… I assume you are speaking into the front side of the mic, not the back or the end.

a USB mic like the Blue Yeti or the Rode NT-USB Versatile Studio-Quality USB Cardioid Condenser Microphone?

Your setup should be good. I don’t know about the sensitivity of the USB mics. The NT1 has a GREAT reputation and a separate interface is usually better. The USB version is probably fine too. The Yeti has lots of nice features but some people get noise through the USB power. I don’t know if it’s really any worse than other USB mics or if I read more complaints because it’s super-popular.

Thanks so much for your answer. I appreciate the guidance more than you know!

I’ll tackle your questions/comments in order, since I haven’t figured out how to clip quotes or copy text in from your post!

Mic cable: My NT1a came with a 10 foot XLR cable. I may try another one to be sure. Everything was new out of the box.

Volume/distance to mic: I have tried shouting, speaking as close as my pop filter allows, removing the pop filter and I still have very small waveforms. I’ve tried standing and even removing my sound shields. If I max out the gain on my solo and shout or clap then I can achieve some clipping sounds. The waveforms start to get that flat top. But even then it’s unpredictable and extremely unsustainable for me to speak at that volume any amount of time. Not to mention it sounds terrible.

Power: Phantom power/power is set on 48v (on my Solo) but it’s harder to tell on the Apogee I borrowed. The mic icon lights up, but no specification if it is the 48v like my solo. There is no led indicator on the actual mic.

Regarding condenser v dynamic Mic: I am familiar with both. I specifically chose the condenser mic for this reason.

Correct usage: Funny you ask because despite my experience, I was speaking into the wrong side of the mic at first but have since corrected and the problems above are all based on the correct direction of my mic.

Still stumped, but partly relieved it’s not something super obvious.

In my experience, recording in the computer (almost) always gives relatively low signal strenght (so, low volume). That is absolutely no problem, as long as the noise floor is very low too (so, when you level up the volume, with compressor or simply normalizing the audio, the noise floor remains acceptable, not disturbing well hearing your voice). (Optional article in Rode site about this: what is signal-to-noise ratio:

Analyzing specifically your audio, it appears that there is some too much undesirable noise:
(The screenshots here are using only the second part of the recording, the original, not leveled)

Normalized (Effects > Normalize) to -3 dB to show the high noise (high SNR ratio).

First thing, then, is to test another ambient to record, turn off air conditioner or pay attention to anything that can be generating that frequency noise. Or, as you say, might be the microphone, only testing alternatives, another USB port of the Mac to connect the interface, or testing a USB mic for example, to detect if that’s a problem. You may also try setting again the knobs of the interface; in mine, a Behringer, the gain is normally good at a little more than 3/4 of the turn (always getting a noise that is constant, so easily goes away with noise reduction).

The thing, the problem, is not a soft voice, as, even with very soft recording, in good ambient and equipment the noise shouldn’t be as high.

Each part of the noise in-between has different frequencies, but we can see (Analyze > plot spectrum) anyway that the frequency here is low, most < 170 Hz).

For testing, I’ve done noise reduction and eliminated frequencies under 160 Hz (as it’s not the main range of a feminin voice, and much of the noise was there), so to have a clearer voice.
(Video showing the procedure I said, 1 min 07 s: - software recorded in the .asf format, VLC opens it.)

The result:

It’s not perfect, as from the beginning the sound capture was not ideal; anyway, I brought it here, since you’ll almost always want to do this when editing your records (noise reduction, normalize, and, not shown here, compression), even when you get it all right! :slight_smile:
02 noise frequency.png
01 stn.png

Hi Antoine!

Thanks so much for your detailed explanation. I really appreciate it. The video was super clear. I have saved it for future reference!

I am recording into my computer directly, so perhaps this is just my reality :slight_smile:

Makes sense, I think, but I need to learn a bit more about the editing and mastering process to follow the rest of your input fully and recreate it myself. I’ll start with the article you linked and go from there.

I recorded this just in my bedroom with a small sound shield (temp setup) just for testing purposes. Far from an ideal recording spot. I am still in the middle of setting up a small studio at home. I will try again in a more ideal recording environment. I guess that’s the first step to trying to improve things!

Thank you so much!

Welcome, evievo; be patient! :slight_smile: I wish you sucess and happinness. Don’t think twice to ask for help, people here know a lot and are willing to help.

Hi Evievo,

I have some of the same kit (NT1A thru 3rd Gen Scarlett Solo into Mac). Those tiny waveforms are odd indeed. You are quite softly spoken I think, but nonetheless should be getting good levels with the #1 gain knob no higher than 3 o’clock on the Solo.

You said the first bit of that sample WAV was mastered, but it sounds to me like it’s from a different source from the others, either the mac’s inbuilt mic or the rear of the NT1A, or conceivably even the apogee’s built-in condenser if you were using the One at the time. Clips 2, 3 and 4 sound much more present and on mic, just at bewilderingly low gain.

A couple of thoughts to add to the mix, applying only to the Scarlett Solo version of your setup cos I know nothing of the Apogee:

Are you able to connect the Scarlett directly to the mac with the supplied cable, or do you need an adapter/hub? (A roundabout way of asking what mac you have.) If it’s a post-2015 macbook with only thunderbolt ports, your choice is either to use the supplied cable with apple’s own thunderbolt to USB A adapter, or a powered hub, or buy a replacement cable to allow you to connect direct (Focusrite recommends getting one of these, no longer than 1.8m.) If you’re currently connecting the Solo to mac through a cheap usb-C to A adapter, that might be one place to start. Perhaps the mac struggles to power peripherals through the adapter, resulting in minuscule gain and the excess noise Antoine mentioned.

You mentioned you could achieve ‘flat-tops’ if you shout/clap. Do you mean flat-tops with clear daylight from the ceiling of the track, like this?
If so, this is properly odd, as if a limiter has been applied before the signal reaches audacity.

All very confusing. If I were you, I’d try swapping out one element at a time to troubleshoot the issue. If you can borrow a condenser mic from a friend, and find you get the same tiny peaks, we can at least rule out a duff NT1A as the problem. For what it’s worth, I think the Scarlett is likely to be the better interface.

Hi GDepot!

Thanks so much for helping me out! So so appreciated! I’ll also tackle your comments from top to bottom :slight_smile:

I have been consistently using the solo at about 3 o’clock gain positioning, a few times closer to 4 o’clock, but I find once I pass the 3 o’clock position, the sound just isn’t quite right.

All samples in the file I linked above were from the Apogee One. However I had the 48v power specifically selected for input so it should not have been the built in condenser.

If you’re interested, here are a few samples from the solo.

I have already taken your advice and switched to USBC to USBC. I was using an apple adapter for USBA to C. I have a 2021 MacBook Air. I had a short cable lying around from my charger. I’ve ordered the one recommended from Amazon Basics going forward in the 0.9m length.

After switching over to USBC to USBC, the flat tops with ‘daylight’ have gone away and now the clap goes right to the top. My ‘loud’ voice also isn’t capped/clipped in the same way. I wish I saved the strange file I had before. Can’t seem to recreate the example I experienced.

Already seems a bit better I think! Maybe that’s just wishful thinking though. I still have to work with the normalization and have a better understanding of which order to apply effects and macros in. I’ll also need to make some presets for myself so I can apply everything at once, consistently.

Also really looking forward to finishing the setup on my small studio space. Hopefully we’ll be ready in the next couple weeks.

I can get my hands on a few other Mics for troubleshooting. I’ll try a few different options!

I can’t express enough how grateful I am for all the shared knowledge! Thank you!!!

small but still passes ACX check when amplified …

I used the free version of couture plugin to attenuate the signal by ~6dB when you’re not speaking.

Hey Trebor!

Thanks so much for the tip and the encouraging update on the ACX pass! :smiley: Will definitely check out this couture plugin tool - thank so much for sending the file back, it really does sound much better after you edited it! I guess it can only go up from here!

Appreciate the help so much!


Condenser microphones that run from Phantom Power as a rule don’t do anything without it. The little round condenser inside the microphone responds very well to your voice, but the signal it makes won’t go down a cable. There’s always a little electronic booster in there to push your voice along and as a rule, it needs phantom power to work. The microphone will not work badly without phantom, it will just fail.

There are News Gathering Condenser Microphones that do not use Phantom Power. What they do use is a little internal battery that has to be periodically changed.

There are really “affordable” “modest” “home microphones” that claim to work from other phantom services than 48 volts, but they usually say in tiny letters at the bottom of the instructions (works best with 48 volts).

USB cables as a rule can be unstable past around 6 feet or 2M. That’s one of the restrictions of USB microphones. You can’t separate them from a noisy computer.

Not so with XLR type microphones. Rock bands use XLR microphones to go the 75 feet from the stage to the audience mixing desk.

Baby steps. Can you make the Solo knobs turn green? That’s going to be step one and has nothing to do with the computer. If you can’t get a green knob with normal spacing and pop and blast filter, you might try oblique positioning and spacing (B).

Leave out the blast filter with this process. Most of the pop and blast noises in your voice go straight in front.

There’s a reason the volume control knobs aren’t labeled. It’s not important. Adjust that knob until it’s flashing green. That’s where it’s supposed to be.

If you get the flashing green knob and still have really tiny blue waves, then something is interfering with the recording. Do you like to use Skype, Zoom, Meetings, or other chat programs? Make sure they are completely shut down before you try to record your voice. Shut down your programs and restart the Mac.

Launch Audacity (by itself) and see if the blue wave sizes improve.

And yes, as above, you should be able to speak loud enough to get a red Solo knob and overload Audacity (100% blue waves.) If you scream (perfectly valid. Do Not Blow) and can’t get the system to overload, then there is something wrong.

It’s very suspicions that as your voice volume increases, the sound quality seems to change. That is not normal.

It’s also not normal that your interface doesn’t call out 48 volt phantom power by name.


I’ll wear my Suzy Sunshine hat for a minute. ACX will not accept an audiobook submission unless they can buy your book on Amazon right now. That requirement is burned into the application.

Also, you can’t read anything from this list (scroll down).

As a fuzzy rule, you need Plot, Settings, and People. They won’t publish a cookbook.

ACX used to offer an audition Quality Control test before you went ahead and read the whole book. That’s the one I failed. “Your sound file is perfect in every way, but you can’t read.”

Good to know.

They don’t do that any more. The first time you find out you can’t read might be when you submit the whole book.

That’s why it’s good to submit a voice test to the forum before you get too far.

I don’t expect this to be a problem if you’re an old hand at commercial reading. We just need to get your microphone sorted.


Hey Koz!

Re: your first message

Thanks for the suggestions and detailed info regarding the condenser mics! As far as I can tell, the setup I have should be more than sufficient for my needs and I think I just need to learn it.

I can get the solo gain dial green, it flashes intermittently from the 3 o’clock position onwards, but then flashes yellow as a warning for clipping sometimes when I have it past 3 'clock.

My audio waveforms are still rather small, but another user said they are formed perfectly fine despite being smaller. Maybe I shouldn’t be comparing my waveforms to other narrators though.

For the oblique spacing, is that normally from a boom arm or something? From a vertical mounting position? Or just placed off the side of the narrator on a mic stand, for example?

My Focusrite solo calls out the 48v power by name, the Apogee did not. But I could select the 48v input on Audacity when using the Apogee which was connected to the XLR cable and NT1A. At any rate, I am back using only the solo interface for now until I can try something else from the studio. Will also test a few other Mics they have lying around.

Re: your continued message

Currently not looking to read just ‘any’ old thing for audible, but more to connect with some Indie authors looking for some more affordable VO work as I learn (still a ways away from even offering my services to others at this point) and work via ACX auditions for now.

I’d hardly call myself an ‘old’ hand at voice over, but I’ve done about a dozen small projects with our partner agency for the brands I work with in my corporate job. They’re the ones who encouraged me to do more after we finished. I’m confident in my reading / narration ability at least… it’s the technical aspect that I find super daunting!

My goal is to get familiar with requirements for ACX/audible, practice the technical aspects and narration until I can consistently generate audio that ‘passes’ then go from there!

Right now, this is more to learn and experiment and see if my narration might be commercially viable in the future for fiction and certain types of non fiction work. Making all kinds of reading materials more accessible to different types of readers is a great project to work on for my interests. I’m an avid fiction reader and I hope some of my reading enjoyment will bleed into the narration as well.

Thanks for the tip about the forum to test my reading abilities. I will definitely do that. Quick question about this test… when I clicked the link, it said:

You should post clean, unprocessed work. Don’t adjust anything before Export. No filters, effects, adjustments, tuning or processing unless otherwise told.

So just record, download and go?

Thanks again! This forum provided SO much more concrete advice than I ever thought I’d receive. Thank you, thank you, thank you!

In some hardware/OS/driver configurations, when a mono recording is made from a stereo-capable source, the signal level can be erroneously cut in half. For those affected users, the solution can be to record in stereo, then delete the unwanted track.

Ha! I can indeed tell. Quite right too. Too much shouting the world.

Things are definitely headed in the right direction, level-wise, assuming the first clip is raw and unamplified. It could probably do with a touch more 1 knob to get your default delivery peaks nearer to -10dB (currently around -15). Very glad the usbc-usbc swap seems to be paying dividends, even if it’s not entirely clear why :slight_smile: The only unwanted noise I can hear in that raw clip is clothes rustle, no buzz at all.

Re the Apogee. Maybe it’s hard to know whether you’re setting the input gain or just the headphone volume, since it has only one knob for everything. Looks like Apogee supplies a free control app, allowing you to set levels etc on the mac. Might be worth a look if you prefer the portability of the One to the Scarlett and don’t already have it installed. I notice its breakout xlr plus instrument cable is rather short and flimsy, mind, so presumably the set-up is a bit cramped.

Here are some unsought booth thoughts from the ‘anything to avoid the nuclear option of draping a duvet over my head’ school of acoustic treatment.

You mentioned various sound shields/cubes. I’ve experimented with these too and for what it’s worth am unconvinced they do much to help - the rear of the mic is insensitive anyway, and sound shields/reflection filters often create unwanted boxy reverbs of their own. Plus they block your view of the copy. The space behind you, i.e. the last surface an echo hits before bouncing into the front of the mic, is the most important place to start. Depending on the layout of the rest of your space, it may even be the only surface you need to treat. Your natural soft delivery is likely well-suited to a less than perfectly treated booth. A heavy, quilted moving blanket hung full length on the cupboards behind me works wonders in my set-up. Obviously, you don’t want to be speaking directly at a bare reflective wall, but better to pile cushions up that wall than have a shield between you and your script imo.

Have fun! I think you’re right to believe your voice is commercially viable.

Hey Jademan!

Thanks for the suggestion! I will 100% give that a try too. Appreciate the tip! :slight_smile:

Hi GDepot,

Thanks for the encouragement. :smiley:

Is 1 knob the gain dial?

Re: interfaces. I am not sure the Apogee One is for me personally. My brother (guitarist and drummer) loves it for travelling but I prefer the solo, I think. The cables are almost too heavy in the compact setup and short cables as you mentioned for the smaller device and make it a bit hard to balance on a flat surface. I will try it with the control app, though. It is certainly more travel friendly than my set up which would be great!

Re: the shields. I have one curved sound shield around the back of the mic but it isn’t so close to the mic itself… I can still read my monitor/iPad/kindle without issue. The shield is more to create a barrier between the mic and the wall my desk is pushed up against. Then I have some sound shield panels on the wall beside me, since it’s just drywall. The shields aren’t ‘cocooning’ the mic or anything. I would love to change to a desk that could have a clamp boom arm in the future, but for now the tabletop stand with the shock mount has to work!

I have definitely learned since coming here that the shields behind you are indeed the most important, so I will make a little clothes rack/shower rod solution to test it out. Our office is rather large so I definitely need to insulate a little corner better for myself. We have tons of moving blankets and comforters from all our university/condo moves. Nothing against duvet tents, but it’s not for me. I find it a bit stifling.

A colleague at the studio recommended something like this for our shared space. A curtain track with moving blankets or room partitions for office sound dampening to make a little cubicle. (attached images).

My brother (who I share the office space with along with his drum set, various amps and sound boards, and guitar collection) will build something a bit more robust in the next weeks I think… maybe a tall room partition with some foam or acoustic pads glued down. We’ll see. He uses the space far more than I do!

Thanks again for all the guidance and support. :slight_smile:

I’ve taken to doing this (with a Focusrite Scarlett 2i2 which the PC logically assumes is a stereo device). It helps.

Oops! Pics didn’t come through.

Hey Koz!

I did the test recording, just a quick read off. This is just from my current bedroom set up. As expected the audio doesn’t pass the noise floor threshold on ACX test, but the peaks and RMS seem ok. Please let me know if I should actually post this sample somewhere else.

I’d love to hear any thoughts/ feedback on the unedited audio. I have my solo gain at just between 3 and 4 o’clock.