Hi,
I’m brand new to this whole editing thing and its alot! I am using 2.3.2 on windows 10. I read on a website that I could use the noisegate plug-in which told me to :
Start with “Level reduction” at -100, “Gate threshold” at 30 and “Attack/Decay” at 75.
Well you can only choose between -96 and -6 on the “Gate threshold” so i’m not sure about that…and setting it and -100 level reduction just takes out the whole recording (no voice sound). I also read on the audacity page to reduce “hiss” I could first use “notch filter” before using “noise reduction” which didn’t do anything…and THEN when I use noise reduction to take out my laptop hiss (which it does well) it adds another annoying sound on my breaths…
Another thing I cannot add this EMI 78 curve in the equalization settings (which suggests is good for cheap mics) http://wiki.audacityteam.org/w/images/4/4c/EQRecord.xml. When I try to import it, nothing is added or changes…
If by echo you mean rever(beration) from the room,
there are de-reverb plugins, but they can only cope with a hint of reverb,
e.g. … https://youtu.be/Zs-5vshUgJ4?t=691
That thing will reduce room reverb to nearly zero because your voice has to exit the space, bounce from the walls and then come back in.
Echo and reverb is terrifically difficult to “filter out” with apps because it’s your own voice arriving at the microphone late. So you are effectively asking software to remove you from yourself.
Noise Gate works by volume control but that has only limited effectiveness because for it to work well, you have to announce perfect volume at all times and nobody can do that. If you miss-adjust it, it will start clipping off pieces of desired speech, and it can have noise pumping effects, too.
Did ACX tell you they didn’t like the reverb in your voice?
If they did, there can be bad news hiding behind their response. They typically only complain about one thing whether you have multiple problems or not.
Can you announce a ten-second test and post it on the forum?
Thank you so much for the replies. I will do that mic/room check tomorrow morning. But in the meantime, how can I change the RMS level? I ran tha ACX check on my recording and it passes all except that.
I haven’t yet sent the completed book over yet, this is my first freelance voiceover job for a client. But i’m trying to learn to edit.
…The Amplify or Normalize effects can adjust the volume and they will are linear adjustments so they will change the RMS, average, peak, etc., by the same dB amount (and they will also bring the background noise up or down proportionally.)
So the usual procedure is to adjust the RMS level first, then use Limiting (which is non-linear) to bring-down the peaks.
I have just literally come across the “audiobook mastering” procedure AFTER I’ve already started editing
Is it too late if I start using that procedure now? Since I didnt save the raw file before touching? I’m thinking all the chapters may come out sounding different qualities otherwise…
I was using You tube tutorials to do the basic “noise reduction” for laptop hiss and normalize for the volume. The RMS normalize makes the voice REALLY loud. I haven’t done the limiter thing yet so I hope that is whats going to bring it down.
The RMS level is a pretty-good indicator of “loudness” (that’s the whole point) and if you’re hitting the ABX target (-18 to -23 dB) it should be about the same as other audio books. (That’s negative dB. )
Just remember, if you are boosting the levels you are also boosting the noise…
I haven’t done the limiter thing yet so I hope that is whats going to bring it down.
Limiting will bring down the peaks and with a “normal recording” that should slightly bring-down the loudness and RMS levels. And, it may make the loudness more-even throughout the recording.
Audiobook Mastering is a suite of tools—a harmonious grouping. The three tools are intended to be used in order because they clean up after each other.
I didnt save the raw file before touching?
It’s very highly recommended that you Export a perfect quality WAV sound file of each raw performance as a backup.
I know it seems like you should save a Project and then edit that Project all the way up to submitting for publication. There are many forum posters who tried that, but had a computer failure and destroyed their work all the way back to announcing it again.
Is it too late if I start using that procedure now?
It’s a little rough to know from multiple time zones and several hours away. That’s why it’s good to post a sound test so we’re all coordinated in what we’re doing.
When you look at that voice test page, tell the browser to reload the page to make sure you get the right instructions. I changed it recently and you should be reading from the page labeled 2019-11-20.
Hi, I have already edited this using Noise reduction to take out the hiss, but as I mentioned earlier it ADDS another sound (don’t know how to describe it) when I take a breath.
[quote=“, post:9, topic:54628”]
It’s concerning that you said the performance got a lot louder when you applied corrections. Low volume can be a serious problem for home readers.
[/quote]
I thought (I read) that ACX read a peak of -3db so I used normalize to do this, but when I use RMS Normalize it goes way beyond that into the red.
Echo/reverb is the one problem it doesn’t have.
The main problems are hiss & USB “mosquito” whine.
Both can be removed, but doing that will generate audible artefacts: the audio will sound processed …
If you want people to pay for your reading, IMO a better mic, (or stand-alone recording device), is required.
You did. It also has an allowable loudness range of -18dB to -23dB, and a maximum noise loudness of -60dB. Two limits and a range and you have to hit all three at the same time.
it (Noise Reduction) ADDS another sound…
Correct. The noise reduction tools are not “free.” They can cause other problems. If you use too much Effect > Noise Reduction, it can add honky, bubbly noises and odd breathing tails.
Can you prepare a fresh sound file according to these instructions? You don’t have to use those exact words. Use the words in the above submission. That silent two seconds is really important to get right. Go down the blue links. They’re very short.
That’s the problem! I’m a VOICE talent not an editing talent. I’m also British so I don’t understand all your (all who reply) jargon/ technical terms and idioms. Lol. Or I’m just not clever. Did you find that voice test I did? It went over 10 served secs so I heard to trim it
Thanks for the compliment I think it’s a combination of the authors style together with my voice reading style. She specifically picked me because my “style” suited her book which is for “normal” people as she quotes, compare to queens English.
This is my first job on Upwork I just want to make a good impression
There’s a raw recording on the bottom of the previous page
Forgive me. Caffeine Deprivation.
Right then. I started with that posting and applied AudioBook Mastering and then very gentle Noise Reduction of the beast (6, 6, 6).
That’s it. I didn’t do anything else.
The post passes ACX Audiobook technical standards and as near as I can tell, should sail straight through Human Quality Control who judges theater and sound quality.
That’s the encyclopedic version. Scroll down to Process.
Noise Reduction is in two passes. Drag-select some pure noise in your reading (that’s what the two seconds of silence is for). Effect > Noise Reduction > Profile. Select the whole piece with the SELECT button on the left. Effect > Noise Reduction: 6, 6, 6 > Enter.
It’s perfect. Is that a slight Irish twist in there?
There is a Noise note. Don’t go Diving for Noise. Set up your listening for pleasant volume during the reading. Then, without touching anything else, roll the cursor back to the silent section and listen. You should hear a barely audible, shshshshssh sound.
Some people roll the show back to the silent section and then crank the speaker volume way up. That’s doesn’t count. That’s not how ACX measures it.
Upwork seems to be a freelancing platform. I can’t see any sound standards or other restrictions.
In my opinion, if you can meet ACX/Audible Audiobook standards, you can publish anywhere. Their numbers are effectively the same as Broadcast, so once you clear that, you should be good to go.
There are some production notes. ACX wants all your chapters to match, so once you start successfully reading, Don’t Change Anything until you finish.
Are you recording with a computer? It gets a little sticky here because maybe you shouldn’t let your machine do auto updates. The forum is full of people whose machines got improved so much they stopped working. Windows 10 posted improvements that made it difficult to use some microphones. There is a Mac OS version update that killed Audacity.
If that machine is your only machine, then it becomes very difficult to isolate it from other apps and settings. Skype for one example, likes to take over your machine and it’s sound settings without telling you.
)
I think it’s a combination of the authors style together with my voice reading style. She specifically picked me because my “style” suited her book which is for “normal” people as she quotes, compare to queens English.