So aside from my issues with plosives and sibilance is my overall sounding better as far as not being boxy or buzzy now?
That’s up to Trebor. He has good ears for that sort of thing.
I can give you a little push with DeEssing.
The current default DeEsser is by Paul L from here.
I use it at these settings after mastering.
You can experiment particularly with that first setting, Threshold or sensitivity. Those settings as posted do work, but not enough.
I’ll apply what I’m learning tonight to tame down the hissing etc in my narration.
Don’t bet the farm on that. There is no known cure for an essy microphone either before or after narration. I would give anything to be able to tell people which towel to throw over the microphone to stop it from doing that.
Essing has very clear symptoms in the Audacity measuring tools, but none of that measuring and testing has resulted in a way to get rid of it.
Read this illustration as increasing pitch left (thunder and earthquakes) to right (air-leaking-hiss and tinkly bells). Louder is up.
That lump on the right is artificial. Your normal voice doesn’t do that. I know you’re saying to yourself now, “Well, just get rid of it.” It didn’t work. If you use the normal sound tools to make that go away, your voice turns into talking-under-a-blanket mud. There has to be voice content analysis algorithms and that makes my head hurt.
As far as applying what I was learning I meant more along the lines of ways of moving my mouth differently i.e. the chewing gum technique and recording with the mic off axis slightly more so than with removal tools.
I hope to hear what Trebor has to say to see if my last sample At least sounds less boxy and less buzzy.
When I make some money I’ll be sure to donate because you guys have been awesome and I truly appreciate you!
The good news …
No boxyness, negligible mains hum/buzz.
The bad news …
The level on seems low, (lower than your other recordings).
Amplifying low-level sound (in Audacity) also increases the hiss noise.
Noise-reduction reduces hiss, but can make the voice computery sounding.
Ideally minimize the hiss in the first place by experimenting with different gain control settings on the mic.
At some gain setting there will be a goldilocks-zone with the best signal-to-noise ratio.
The sibilance is too a bit too strong. To desibilize with Paul-L’s de-esser …
Step #1: measure the RMS of your audio, (the ACX check tool tells you that).
Step #2: Apply Paul-L’s de-esser with these settings: set the threshold to the RMS you measured, plus 1…
(If too much de-ess is applied it will sound like you’re lisping).
I don’t think it is possible to reduce sibilance other than by processing the audio.
The off-axis trick helps reduce mic-popping plosives, but not sibilance.
Thank you so much! I’ve been tinkering with gain for like 2 hours and I think I found one that is a pretty good sweet spot! I am going to go ahead with recording my ebook with these settings and I’ll be back to let you all know if ACX accepts it. My fingers are so crossed! lol Wish me luck!
If you do get bumped, please post the whole rejection note. They usually go to some trouble explaining what happened and possible solutions. That’s how we know what they’re thinking enough to tell forum posters what to expect.
So I just got news back on the sample I submitted to ACX. They said it fails due to a buzzing sound in the noise floor which I cannot even hear. I know you guys heard it in earlier samples. Since that is the only thing failing me I submitted a new sample in hopes it is gone. Can you all see if you hear it in this snippet please?
If so I need to work incredibly fast to figure out the culprit yet again though I have a feeling it’s gone now.
I don’t hear a buzz in the noise floor.
Is that sample taken from your original Audacity Project, or from the MP3 file that you sent to ACX?
Please post another short sample but export as a WAV file rather than MP3.
I thought that too. I wondered if that was the result of encoding to MP3 twice. On the other hand, maybe that’s what ACX were complaining about, but didn’t have a check box on their system for logging that fault.
Let me get a wav sample. I actually did paste in room tone to cover areas with mouth noises or weird breath sounds in between sentences etc.
I used the copy punch paste feature to put the room tone in those areas. Ok wav snippet attached. I sure hope these will work because I am over half way through a 60,000 word novel for FindAwayVoices. However I think I’d improved A LOT from the original sample I sent ACX which I cannot find now, I don’t remember which sample I sent them which is really bad cause now I can’t compare for improvements.
They took a long time to get back to me and so with an author waiting I just went with recording and hoped for the best.
I put room tone over almost every breath sound unless they’re in the middle of words in a sentence. I don’t like hearing them before new sentences so I doubt listeners will either. Is there a better way to deal with that?
The cross-fade time is too short causing the ends of words to be clipped off,
which is giving the game away that that technique has been used.
I’d use at least 100ms, (0.1 seconds), cross-fade time to hide the joins.
I’m lost on that. I’m not hearing the ends of my words being chopped off. It’s weird too cause the author has proofed this and she didn’t notice it either. I guess I need to read up on the cross fade cause I’ve never heard of it and don’t know how to use it.