HELP! Same set up over two days, but different audio quality!

Hi! Please help, I cannot figure this out for the life of me. I’m a first time narrator for ACX, but I have voiceover and (film) editing experience.

I painstakingly recorded my first day of audio for a book, set all of the specs, etc. So far, I think it sounds pretty good!

Then, I did the same thing the next day…and the sound is practically unusable!!! I’m including examples. Again, SAME set up, SAME specs. And, to test it further, garageband did the same thing as well! Perfectly fine audio one day, and then today it’s tinny and sounds like I’m recording from my cell phone.

I thought, could I possibly have been recording from my computer? But I don’t see how that would work as A) garageband ALWAYS asks me if I want to use my samson meteor mic instead of the built in one, and B) I can hear my voice in my headphones as I speak through the mic, and know the sound is coming through the mic instead of the laptop.

The ONLY difference between today and yesterday is this: I was fiddling around and looking at forums to try and figure out why I couldn’t hear myself through my headphones while I was recording (in audacity only). Yesterday all I could find was having it play in my headphones with a delay, which is obviously very distracting when narrating. I feel very strongly about being able to hear myself when recording, but couldn’t figure it out. I don’t know how, but one of the times I opened audacity today, it just worked! I could hear my voice in real time, whereas yesterday I had to record without hearing anything through my headphones. Now my audio is horrible. But if this is the key, how would that explain that garageband also has the same result??

I am now in deep trouble as I have a deadline tomorrow for a completed children’s audiobook (very last minute turnaround needed). Had things not gone well yesterday, I wouldn’t have waited. Now, I don’t know what to do. PLEASE help.

I feel very strongly about being able to hear myself when recording

Strong enough to miss your deadline?

As a general rule, Zero Latency Monitoring is only available from the microphone or interface, not the computer.

770×800 41.2 KB

There are specialist ways of doing low latency monitoring, but they’re not done by messing with settings. They generally involve intentionally installing software.

You can inspect Apple > System Preferences > Sound and make sure actual physical things are selected for input and output audio. Built-In Output or Headphone Output, for example, and not SoundFlower, which is a “fake” device for special application.

Then make sure Audacity is set for similar devices. That would be the little Microphone and Speaker icons in the device toolbar.



I expect that to give you clean, clear audio recording (depending on your environment), but I don’t expect you to be able to hear yourself without an echo.

The meteor has a headphone connection.



Koz

Audacity has Audiobook tools and techniques, but it may already be too late if you’re close to deadline. If you get rejected by ACX, post back.

Koz

Hi Koz, thank you for such a quick response. I don’t think I explained my situation correctly, I’ll try again! …I don’t know all the necessary terminology.

(I have been using garageband for commercial voiceover auditions, and there is zero latency there with the mic. I also triple checked online and found manny different sources confirming the headphone jack in the meteor allows for zero latency monitoring, so I’m not sure what I’m doing wrong.)

However, this is an issue for after this audiobook has been finished. What I cannot figure out for the life of me is why these two recordings sound so different, under the same exact conditions and specs. The only difference between these recordings is that they were on two separate days, and that somehow for recording with poor audio quality, I DID get zero latency monitoring to work in audacity.

I’m having trouble attaching so here are links via dropbox:

Day One: https://www.dropbox.com/s/fcrqbjp4dniku8y/essential%20oils%20acx%20spec%20sample.wav?dl=0

Day Two (bad audio): https://www.dropbox.com/s/qxvqt247aetuf8c/Bad%20audio.wav?dl=0


I am going to re-record of course, but need to fix this problem! Thanks for the help, much appreciated.

What are your settings in the device toolbar?
What are your settings in “System Preferences > Sound > Input”?

I would say with ~90% certainty that your “bad” recording was not using the Samson Mic, but instead is using the built in microphone in your laptop. The change in timbre and “shower stall sound” both point to that being the cause.

But also your “good” recording has a fairly nasty case of the “frying mosquitoes” aka “Yeti curse” a high-pitched whine that results from the USB data communication signals finding their way into sensitive microphone pre-amp. The “frying mosquitoes” are notably missing from the “bad” recording.

You are monitoring the recording via the headphone jack on the Samson itself, correct? As Koz points out that is generally the only way you are going to get zero latency monitoring. And I’m sure you were hearing what the Samson was producing, but the laptop wasn’t listening.

If you have any doubt about which microphone is recording, do the “scratch test”, scratch the microphone wind-screen with your fingernail while watching the recording meters. Scratching the wrong microphone probably won’t register, where as scratching the right one will probably run the meters into the red.

http://kozco.com/tech/audacity/clips/ScratchTest2.wav

You will have to deal with the USB noise in your “good” microphone before you will have an acceptable submission.There are a number of things you can try which may or may not help: shorter, higher quality USB cable, adding a powered USB hub between the microphone and the computer or adding a USB isolator between the computer and the microphone. There is also a comb filter plugin that Koz can post the link to, which will help but it will also have some effect on the sound of your voice.

Thanks so for the help, flynwill. I’ll do the scratch test and hopefully it was an oversight on my part with everything else I was trying to do. I agree completely with the Yeti Curse. I ran a plug in I found that had good feedback called MosquitoKiller4, but listening back after getting some distance from it, you’re right, it isn’t good enough.

Which do you suggest I try first of the solutions you offered? I’d like to make a long term fix if possible, but if something can be done short term for this short deadline, it would be a lifesaver. Koz, could you please send that plugin flynwill referenced?

This is definitely on me for agreeing to such a short turnaround time, but I thought I’d worked out most of the kinks and the audiobook is only about 14 minutes total. Well, learning from my mistakes here.

Also, it seems I should probably invest in a higher quality microphone? This was actually the higher quality microphone recommended to me for my commercial VO work, but I’m learning probably isn’t good enough for a proper home studio. If at all possible to make this mic work for a few months until I can properly invest, that would also be very helpful.

Thank you so much guys for taking the time, Sarah

but instead is using the built in microphone in your laptop.

What he said. Do you have and use Skype or Chat? Those two can reassign sound pathways and processes and not tell you.

I ran a plug in I found that had good feedback called MosquitoKiller4

Too late. That’s the plugin designed (I don’t remember by whom) to suppress commonly experienced whine. As near as I remember, it’s the only way to convert what you have into something presentable. However, it works by surgically removing some musical tones from your voice. It creates damage that happens, on the whole, to be desirable. Far better, of course, to not have whine in the first place.

There are stressful ways to get rid of the whine forever: Keep changing computers until it goes away. Change to a much more expensive USB microphone. Stop using USB microphones.

The Yeti microphone (hence, “The Yeti Curse”) has no filtering, processing or effort to suppress this effect. Lots of people on the forum complaining about that whine. I don’t think we’ve ever had complaints about a Yeti Pro®. Pay attention. I’ve seen plain Yetis advertised as “Professional Yeti.” Yeti Pro® is a registered product that has the XLR connection on the bottom. As a side issue of the extra tricks it can do, it’s much less susceptible to the whine problem. It’s also twice the price of a Yeti.

In My Opinion, you should be using what you have until you can do it with no side issues or obvious sound damage. “I think I can do better” doesn’t count. Also In My Opinion, you can do a world of good by improving the environment. Are you using the Meteor on a plain table or desk? You can improve the sound by padding the table (slap and comb filtering effects) and using the book and towel technique (table and floor rumble and noises).

I’m a fan of furniture moving blankets. Great things, those blankets.

How you sound to you is almost secondary to what you sound like to customers and clients. If you’re presentable to a majority of people, that’s what matters.

I mastered a portion of your work for ACX.




That’s not straight mastering. I had to add Mosquito Killer 4, Noise Reduction and De-Essing.

Still, it took me about 30 seconds and you should be good to submit.

Koz

Other clients may differ, but ACX would rather you submit in one track mono, not two track stereo. My posting is configured for mono (one blue wave).

Use the drop-down menu on the left of your stereo track > Split to Mono. [X] delete one of the two tracks.

I can listen to a story in that voice. You are warned that ACX does not like changes in the middle of the book. Once you start, you should stay with one microphone and environment.

Koz

Let us know if you want to continue this way.

Koz

It occurs to me to ask, did you already read the whole book? That’s a common New User error. Read the whole book with technical problems and only then check if the work is OK. If you hit particularly evil errors, you get to read it all again.

Koz

This has been so helpful, thank you thank you.

That’s not straight mastering. I had to add Mosquito Killer 4, Noise Reduction and De-Essing.

Kozikowski, could you walk me through the exact steps you used for Mosquito Killer 4, Noise Reduction, and De-Essing? Are there specific specs or can I just run them? I already had the Equalization/RMS/Limiter specs running as per ACX requirements and also use ACX check. It had passed with the whining, though I knew it needed fixing.

Yes, this is exactly how I would like to continue, and I’ll just re-record one day of work. You guys were right about the in computer recording. Are these the best steps to take for future audiobooks?

This is my current setup:

Screen Shot 2018-10-06 at 1.30.31 PM.png547×664 311 KB

Also for clarifcation, yes I use mono! And also I have the Samson meteor and was told to get it instead of the yeti by my professor. (Actor here )

Also! Do you recommend applying plug ins before or after manually removing breaths with room tone? This is the only other editing I have been doing besides plug ins. Are all my bases covered?

No, I’ve only recorded the first 15 of Essential Oils. As for the children’s book, it’s only 14 minutes long so re-recording isn’t a problem.

I can do all those things, but you caught me one foot out the door.

Soon to be followed by the other one.

When you quote someone, select it and click the fat quote marks above the forum text panel. I patched one quote so you can see what it looks like.

As we go.

Koz

Just figured that out too, thanks Okay fantastic, what I’ll do now is re-record the audio and then edit and master it later tonight when I get back from a job, so any other advice you have, when you get a chance, would be really appreciated!

Overall Room Tone is quieter post mastering, so you may find it easier to hide breaths and other mouth noises.

I think I’m making some unwarranted assumptions. I picked Oils with Noise from your Dropbox example postings as “raw” and with no processing. That’s not true, is it? I started to dig and discovered signatures from some of the effects.

~~

How I did it.

Effect > Mosquito-Killer4 > 8 mosquitos > OK.

Mastering 4 just as it is.

https://forum.audacityteam.org/t/audiobook-mastering-version-4/45908/1

Effect > Noise Reduction.

Right there we run into the first problem. Your Dropbox-1 posting has two different room tones. One fake which makes me think you made it and stuck it on the beginning. The real, higher room tone with the whine in it only starts about a half-second before your first word.

Quiet Quiet Quiet Quiet FFFFFFFFFFF “Copyright 2018…”

That’s dangerous because ACX Check has terrible accuracy if you don’t give it at least a half-second of “real life” room tone, and it’s not accurate at all if you have all that “fake” silence. ACX Check uses the quietest half-second it can find. ACX Human Quality Control (not the robot) is going to catch that noise level shift pretty quickly.

So I cut off all the fake silence and extended the louder room tone (copy/paste).

Yes, it does pass, but just barely. 62dB out of 60dB limit

Drag-select a chunk of that new room tone noise. Effect > Noise Reduction > Profile.
Select the whole track by clicking just right of the up arrow.

Effect > Noise Reduction > 6, 6, 6 > OK.

That’s technically not needed, but it sound better because the shshshshsh sound between words is quieter.

Then it’s up to you. Listen critically before and after applying Effect > DeEsser at the default values.

You’re listening for overly crisp and harsh SS sounds. My joke is the “ice pick in the ear” sound, to be avoided. Microphone makers think that sound makes it more “professional.” It doesn’t.


I see your studio. You’re much further along than normal New Users. That explains why I was missing some of the errors when I listened the first time.

Is there inside padding on the top , bottom and back? Is the microphone sitting on the table? You may still need that towel and book thing. Anything that fits. Half size from the illustration. If you don’t do that, I bet every time you rap on the table it gets into the show. Or worse, somebody in the apartment downstairs.

One oddity outside the box. Don’t put a plain wall straight behind you. The sensitive part of the microphone is towards your face and the wall behind you. Perfect for picking up room noises. So your back should be pointed toward a wall at an angle or pointed at a corner. This is hard to write. Rotate the desk and your chair slightly so the microphone is not aiming straight at a wall if you weren’t there. I have no idea what the rest of the room looks like.

Or your could soundproof the wall. Hang one of those furniture moving pads behind you. My personal favorite was the room at work with stacks of cardboard boxes. Perfect soundproofing.


Record a test rather than re-purposing an existing recording. 20 second mono WAV files will fit on the forum.

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

Koz

World’s worst illustration. Top View. Microphone and box pointed at a wall. Doesn’t have to be that extreme a tilt, but any angle other than straight-on is good.



Koz

To answer your questions several posts ago. I would say that yes you would be better off spending money on better kit vs on possible fixes for the mosquitoes with the current microphone. Folks seem to have better luck with separate USB interface with microphone inputs (there are a lot to choose from) and a separate XLR connected microphone (where again you have a ton of choices).

In the meantime if you can get work out the door using the mosquito killer plugins go for it.

Adding to what Koz was mentioning about not having the wall directly behind you. Be aware that Corners are even worse than a flat wall, because no matter the angle between you and the corner, any sound from you comes directly back to you. (google “corner reflector”)

Hanging a moving blanket behind you works really well and can be easy to set up an knock down.