I am using Audacity 2.0.6 Windows in a closet. I use this formula when I clean it up:
Back in the equalizer: Select Curve > LF_rolloff_for_speech
It should look like the picture attachment. Don’t forget to push the Length all the way up.
OK.
Effect > Normalize > [X]Normalize to -3 [X]Remove DC
No it doesn’t. You have a hum or buzz in there that’s wrecking your presentation. Post 10 seconds of mono wave from the original shoot. No filters or effects. Pick the first ten seconds from that clip.
I need to get to a larger machine, but just first level “quicky” analysis fails ACX Noise.
Darn–I think that’s my heater that’s on. Let me try recording silence in mono when I know that it is on and when I know that it is off. If this is the case, then I just have to make sure it is not on when I press record–I hope!
I was going to “Get You Out Of Jail” with a notch filter, but that didn’t work. Past the obvious single tone that fan is making, it’s also making other rushing noises that are not so easy to suppress. Besides, the notch filter can distort your normal speaking tones.
That leaves Noise Removal. Noise Removal is a violin. It has to be played. You don’t just push a button and all the noise goes away. It affects both the noise and the performance and it’s up to you to balance the two.
It occurs to me to ask why we’re doing this. Do you intend to publish to audiobook or other commercial work? If not, what you have there is probably OK. ACX has the problem that all of its thousands of publications have to be top quality and match each other. If you’re doing a private blog, you don’t have to worry about hum pitches and noise floor. Your posting appears to be close at -55 or so. ACX has to be better than -60.
I am trying to get past ACX. When I started all this, I didn’t realize I was going to have to have a mega recording studio with thousands of dollars of equipment.
I made that noise clip intentionally bad so you can hear what I’m hearing. It’s the sudden recognition that it’s probably your goldfish bowl pump. It’s not that bad in real life.
I didn’t realize I was going to have to have a mega recording studio with thousands of dollars of equipment.
You don’t. But most people think you can plunk a microphone down in the middle of a noisy apartment and make ACX conformance. You can rarely get Noise, RMS (Loudness), and Peak (overload) in the same clip if you do that.
As a test, I set up a simple rock band microphone, small mixer and my Mac in my third bedroom and shot a sound test. With the exception of my Hum Generator Monster in the Attic, and very gentle corrections I created what we perceive to be a perfectly good ACX compliant clip. It’s not rocket surgery, but the ACX guidelines are broadcast/pro quality. They’re not fooling around.
And you are close. I measured everything else in your clip and it sails straight through. Noise is a killer.
You reversed those two clips, right? The Heater clip is clean and the No Heater clip has significant hum tones at 60 and 120.
Do a test clip without the heater.
Room Tone
2015-02-28
Room Tone is the sound your room makes without your performance.
ACX guidelines tell us to include a room tone segment in submissions and tests. Room Tone is also good idea when you post on the forum, particularly if you have questions about voice quality and ACX Audiobook conformance.
Set up for a normal performance. Everything normal as if you were going to launch straight into your monolog for a recording. Stop moving and hold your breath for two to three seconds. I mean stop dead. Not start checking your Twitter or Facebook feed or fidgeting in your seat. We can hear that.
You can slate it, too.
“Room Tone.” [---------- silence ---------] “This is a voice test. I will now talk randomly for seven seconds…etc.”
You can post a ten second WAV (not MP3) sound clip at 44100, 16-bit, mono on the forum.
…or on your dropbox account. Just noise doesn’t tell us enough. As I posted somewhere else, you can make any microphone sound like trash if you turn it up enough