Will it pass ACX check? It does based on my Audacity plug in (noise floor, RMS loudness, peaks, silence at beginning and end).
Specifically I have had trouble punch pasting room noise into areas of the audio where there was breathing etc. It has been difficult to find a piece of room noise that will work well (some with a weird hum, some close to noise floor). I found one in recording, Normalized loudness to -75dB, and used that. It sounds good to me, but not sure about to those more experienced. Does it sound natural? Will it pass the human listener quality test?
Also, my father obviously isn’t a professional narrator, but will his performance meet ACX standards?
Any other suggestions or recommendations are welcome.
I can listen to a story in that voice. ACX is not in love with BBC/Hollywood voices. Their goal is somebody telling you a story over cups of tea. I used to say “In The Kitchen,” but I don’t do that any more because a kitchen is a terrible place to record sound.
One of their primary goals is No Distractions. That gives you Sarah Vowell who has the exact opposite of an announcer voice, but can tell a story. I have all of her CDs.
You can have editing problems. ACX doesn’t much like you inserting dead silence into the performance. As near as I can tell, it’s an indicator that you have been messing with the sound and chances are good you’re not a good editor and are likely to damage the voice. So that’s the pasting Room Tone into places instead of Generate > Silence.
Nobody said that was easy, either.
Do you have good quality WAV or Audacity Project files of the performance before editing? Edit Masters? This one isn’t obvious. Many new producers create the ACX MP3 submission files and that’s it. If they fail, even for a relatively simple and easily correctable error, it’s back to the microphone. You can’t easily fix an MP3 file.
Another common error is pile everything into one Project File. It’s clearly posted in the ACX standards that you do everything chapter by chapter and they all have to match. One computer crash and it’s back to the microphone.
I wouldn’t mind you posting a WAV file of some of that work currently appearing on audio.com. It won’t let me download any of it. If it really is a mono file (one blue wave) you should be able to post about 20 seconds of performance on the forum.
Thanks for the input. Please see below for small WAV file from the same chapter. Also replies to some of your points.
Great to hear about voice standards.
For the “No distractions” and editing problems points, I hope I’m hitting those. Certainly questionable in terms of pasting in Room tone in lots of places, but I think it sounds fairly natural (not sure though).
I definitely have all of the original Audacity project files (and any revisions along the way editing), and will keep in mind the separate chapters requirements and keep everything separate.
I can wear my obsessive hat for a minute. This is one of the reasons I wanted to hand-hold the file.
That beep in the middle is me boosting the Room Tone-Background Noise between words. What’s supposed to happen is spring rain in the trees shshshsh or fffffffff pink noise. That’s natural microphone or recording noises. What you have is digital radiation, screen crosstalk, or software trash.
It’s low enough volume that nobody cares, don’t worry about it, but it really shouldn’t be there. I did a test once where I tried to read a script from a tablet. The tablet made noises in the performance.
The room tone you’ve added is not typical, it looks like a ~3ms sample looped, so it has harmonics (the horizontal parallel lines on spectrogram) rather than noise, which is random, (without repetition) .
Your mic has resonance between 11kHz & 13kHz, this adds a gritty harsh sound.
Dynamic EQ to cut back 11kHz-13kHz will reduce it. TDR Nova is a free plugin which can do that.
dynamic-EQ only kicking-in when necessary is a better solution.
The EQ sounds a bit muffled: lacks mids & highs … TDR Prism | Tokyo Dawn Records (free plugin)
and the compression looks odd: each phrase starts loud, and I can hear levels pumping.
[ Possibly due to Audacity’s slow attack/release compressor: The free version TDR Nova is way better than Audacity’s compressor ].
Any beginner tutorials for TDR Prism you recommend? I’m wary of jumping into a new program seeing as I’m only halfway able to use Audacity well.
Would I go back to the original recording files, put them in TDR, and then master there? Is there a specific Macro setup that would work in TDR like there is in Audacity? Or what would the work flow look like?
TDR Prism is a free VST3 plugin that works in Audcaity3
You can load a reference track into TDR Prism to compare with yours, see … https://youtu.be/tMzQVOfNVbo?&t=467, but you have to manually EQ your track to match.
[ There are TDR plugin$ which automatically match EQ ].
TDR Prism does not come with any reference tracks. You could use an audiobook read by a male as a reference, or this noise which matches a generic male voice …
Without question the worst thing that ACX did is kill their human evaluated test submission. That’s the one I failed. My reviewer said my sound file was terrific, but my voice wasn’t. I was too noisy and distorted for audiobook reading.
Ever since the the pandemic, they found they could get away without the evaluation forever.
Your choices are submit the whole book finished…or not.
OK, so here we are well over a week messing with it.
How did you record the original voice? I think I mentioned up a bit that I can hear odd digital and process distortions and some software “helping you out” with the quality.
There is a Give Up Point and record it again with better equipment and environment.
That’s a sound studio I whipped up for people in the building to walk into and perform voice tracks for a movie. It takes me about fifteen minutes to set it up. I used to borrow one of the film editing rooms from the editorial department.
Much later I designed the Kitchen Table Studio.
These blankets are all furniture moving pads. They kill echoes and make anybody sound terrific. Note the folded over one on the floor and the desk.
I’m not that much of a fan of recording on the computer. It’s too easy to get the damage that you got.
I have a Zoom H1n recorder, several Olympus recorders and I can record sound on my iPhone. They will all record top quality sound. All outside the computer. Stay away from MP3 recorders.
Setup:
Hardware: AT2020 with pop filter on a vibration shockmount. Long USB C cable. Wired ear buds (two actually with a diverter so my dad and I can both hear).
Environment:
In a small walk in closet with blankets on all the walls. Potentially not thick enough I suppose. The cord from the mic was run outside under the door and to a laptop.
I’d certainly be willing to start over, but not sure what I’d change to improve. I did not have anything on the floor, so could try that I suppose.
Also of note in review: I realized the macro I had been using for editing had included a compressor, but not Graphic EQ. I went back and tried some with the original Audacity ACX mastering macro (with graphic EQ and no compressor), but I can’t really tell the difference.
Do you recommend using graphic EQ (I’m unsure of what curve would be best)? What about compressor?
Cable length of 10feet. Somewhat required to get out of the room and to the laptop.
It is certainly possible that the laptop makes some small amount of noise during operation. I have had it plugged in sometimes but not always. I figured since it was outside the door it shouldn’t be a problem, but I can always have it plugged in moving forward.
One thought that I had was the light source in the room creating some noise the mic could pick up, so attached is a recording using a headlamp rather than the light in the closet. Does it still have the same issues?
You’re oozing into the right mindset. Your last light-free sample has a good illustration of the problem.
I boosted the noise after “1953 Cessna” and then again after “Most Folks’ Standards.”
Note they’re reeelly different. Something got it in it’s craw to jump in and help you in the second gap.
If you’re just plain listening on large, fuzzy headphones, all you will notice is the shuffle, shift, FFF normal environment and microphone noise will suddenly vanish. If whatever did that substituted silence we wouldn’t be here. But it didn’t. It leaked trash. There’s some other trash as well, but that’s the one I noticed first.
I know you’re going to look up expectantly and say, “So how do we fix that?”
I’ve never met that problem before. I’d be carefully packing up the computer and recording on something else.
I don’t know that I have a voice sample on that. It’s a Zoom H1n standalone recorder. That picture was taken in my outside studio. That lighting is coming from the white building next door.
Other than that, I want to find out more about that Cessna.
Actually, you may have an AT202USB+. That’s the one with a headphone connection and the native USB service.
What you’re supposed to do is wear your big, fuzzy, sealed headphones and listen to yourself closely to keep yourself from fading in and out as you read—like you do in that final phrase.
“Turns Out” is clear and robust whereas “1953” takes leaning over to hear clearly. That may also be what’s driving your little hidden digital helper nuts. It can’t tell when you’re being Hollywood Expressive, or just getting tired.
If your voice was more even, it might make post production editing easier.
I guess I’m not savvy enough to use the plug in. Got it installed, but in order to open sonible I have to have a track selected, then once sonible is open it says to play the track, but I am unable to play the track no matter key I press or place on the display I click.
I see. Not sure how much of that I’ll actually be able to implement, with my father not being around all that often to practice/ record. I’ll keep it in mind though.
