so im currently recording a aduio book for audible and i have to meet the ACX expectations and im using the RMS Normalizer and the Limiter to set the levels to pass but they only get close (i need em to be exact for ex : i set the rms norm. to -20 db → its 26,…) im using the latest version of audacity and my rms needs to between 23dB and -18dB, my peak lvl has to be -3db and max -60db background noise.
PS: procedure → 1. record 2. set rms lvl 3.limit to -3db result–> not exactly the values i need ;(
I haven’t gone around and cleaned up the older versions, yet. SetRMS is the older program. The new Program is RMS Normalize. You also need to know the default RMS Normalize isn’t good for audiobooks. Change the setting from -18dB to -20.
The updated and checked AudioBook Mastering version 4 is here.
This document also has a separate paper on noise for when you get that far. There are several Noise updates and programs missing from the older processes. We can recover from some broken microphones now.
One item stressed multiple times down the paper is Export a WAV copy of your raw reading in a safe place. There is a temptation for the New Reader to read a chapter, process, clean it, and submit without ever making backups. The first time something goes wrong, you’ll be reading it again from the beginning.
Let me know if you get stuck on this new process. I tried to make it step by step and relatively simple.
There are also ways to check your work as you go to see where you went off the rails. I didn’t include them to keep the paper short. If everything goes well, nobody wants to hear painful details about audio theory.
i downloaded the new version of it and it does set the rms to the value i want it to but if i use the limiter now to set it to -3 the rms changes how do i set both to the values i need them to be without the one affecting the other ?
One oddball observation. Are you in Mono (one blue wave)? Some of the correction tools can act wonky if you’re in stereo (two blue waves). It’s not a shootin’ offense, but ACX would really rather you be in mono and stay that way through the whole show. They will accept stereo if you insist.
okay so i’m using the latest ver of audacity (2.1.3) and have it set to stereo.
I set the RMS to -20db and then i set the limiter to -3db if i do that the results (with the acx check tool) are : peak lvl : -3.5db rms lvl : -27.1db (fail) and noisefloor : (-290.0db) (i have several audio tracks and it measures for everyone of them and they all have different values but are quit similar (the RMS lvl is a bit higher and a bit lower on some, like 29.2, 30.9 and 24.7) to the values above)) and i haven’t used low rollof will try that out now.
I don’t know where to go with this. Some of your results are nonsense numbers.
how i do the equalization rollof thing
I can talk you through that, but you can’t do these steps out of order. Low Rolloff is intended to go before the other steps. It fixes some problems that can harm the other tools.
I think you should stop reading until we figure out what’s happening.
How are you reading? What kind of microphone? When you get to the end of a chapter or wherever your break point is, do you Save an Audacity Project or Export a WAV sound file for protection? It’s dangerous to dive right into filters and effects without saving a clean protection file.
We can’t take filters or effects out of a reading. If something becomes damaged and you don’t have a backup, you may have to read it again.
so 29.2, 30.9 and 24.7 = -29.2 -30.9 and -24.7 (forgot the -)
i do save but the last 2 recordings were saved after i did normalize and limiter (normally i do it the other way around but sadly didnt do it this time …) if i have to ill record it anew tho.
my microphone : Behringer b-2
i dont exactly understand what you mean with drop out but if you mean you have to go now or have no time etc. then okay but i would highly appreciate it, if you could help me as soon as possible because i need to deliver it in 4 days
The elves on the forum are all part-time, volunteer helpers. Every so often, Real Life pops up and we have to stop.
If you found a way to process your work without going through the diagnostic analysis craziness, then you should do it that way. Fair warning ACX Technical Conformance is only the first half of the ordeal. After you get the blue waves settled, you have to submit to Human Quality Control and that’s where they do theatrical and sound quality inspection.
For example, nobody is going to pass your work if you sound rough, spacey or like a bad cellphone.
Another warning, They take a very dim view of you changing your sound quality in the middle of a book. Once you start a book you need to do it the same way to the end. This just kills people who insist on millions of corrections or filters at each chapter. Even worse, the readers correcting their presentations word by word. They’ll be in the retirement home by the end of the book.
Have you contacted them about submitting sound samples? There’s a standing joke about the Audacity poster we got all the way through final audiobook acceptance and they screwed up the number of seconds of silence at each chapter introduction. You do have to pay attention.
So in the mean time, we should find out why Audacity is failing.
Neither my C-1 nor your B-2 will plug directly into a computer. I have a Behringer U-Phoria UM2 computer adapter. How did you do it?
Low Rolloff for Speech is a standard, built-in Equalizer preset in Audacity 2.1.3. This is an excerpt from the instructions.
Select the work to be corrected or a whole performance by clicking just above MUTE at the left of the timeline.
Effect > Equalization > Select Curve: Low rolloff for speech, Length of Filter: about 5000 > OK.
This is the whole document.
Using the filter is two sentences. Explaining what you’re doing and why is the rest of the page.
first of all i would like to thank you once again for helping me !
im recording for audible and already once did an audiobook which passed the ACX test etc. and was uploaded so i “don’t” have to worry about that i think the problem with audacity was, that i had to export it so it would be on one track instead of 35 tracks (beacause i record passage for passage) (when i did that it worked on audacity). My mic is hooked up to a pronomic m-502.