I am using Audacity 2.3.0 on Windows 10. I am working with recorded audiobook file and need to change settings to pass Audible requirements. In running an app that test the audio it says that the “Noise Floor: 0.022814 (-32.8 dB) << Exceeds Noise Floor”. Keeping in mind that I am hack at this, how can I manipulate the recording to meet the requirement? I do have the ability to re-record as the VO talent is not available at this time.
First, the good news. If the presentation sounds OK, the noise is probably not as bad as -32dB. That noise level would cause you to run from the room holding your ears.
Now, the not so good news. The probable reason ACX Check failed was it couldn’t find any good quality background sound (Room Tone) to measure. When that happens, it picks the quietest thing in the presentation and measures that. So that might be natural breathing or other sound part of the show—generally much louder than the actual background sound.
Now the really bad news. Room Tone is part of the ACX Submission Requirement. It’s not optional and you can’t make it later because machine-made background “sounds funny” and is easily detected in Quality Control.
Has the voice artist ever read for audiobooks before? This kind of thing comes up immediately on the first submission.
You should probably describe the job. Who hired whom to do what and how? It’s possible you got the client submission and the studio masters are still available.
This is one of the other posters on the forum. If you analyze this clip, you find it fits inside the ACX requirements. It also has a nice chunk of well-behaved background sound at the beginning.
First: I meant to say the VO talent is NOT available to rerecord.
Second: THANK YOU. I understand what you are saying about background noise and am sure that is the heart of the problem. I’ve been zeroing out back ground noise in some cases as it was easy to do and did not realize the importance of not doing that. I will try laying back in some background from unedited tracks to give it the proper base level. Unfortunately it may be a couple of days before I get to work with it, but I will get back to you next week and let you know.
Pumping Room Tone isn’t good, either. That may fail Human Quality Control.
The submission should be a natural reading with maybe gentle noise reduction of the beast (6, 6, 6) over the whole thing if required. If you have any of the other “standard” noise problems like The Yeti Curse, some of them have pre-baked correction solutions, too.
It is Very Important that you save perfect quality WAV copies of the original reading in a safe place against just such problems as you’re having. The original reading should have been delivered to you as WAV as well and the voice artist should have local WAV copies of their performance. You can’t easily edit MP3 sound files without causing sound damage. ACX demands MP3 submissions in super high quality (192-Mono) because of their production system and local storage. You should be in WAV the whole way up to submission.
I think the premium is directly related to the prominence of the author. I have none. And yes, my voice pretty much sucks. I’ve sampled a few times and the “focus group” response is not great…and I’m honest enough to admit people really wouldn’t want to listen to me. And finally, the main character is female. I’m not.
Again, thanks. I’m sure I’m not done here but need to employ the info you’ve given and see what I can do. You have been greatly helpful and much appreciated. I will come back and update even if all goes well. Don’t know what I would have done if not for this site.