Attached is a raw clip using a Toshiba TECRA A50-A, 64-bit, running Windows 7. Boom-mounted Blue Yeti connected directly to the computer. Recording done in a Whisper Room SE2000 series, with computer outside the room. Whisper room located in a totally-enclosed, windowless internal office on 9th floor of an office building. Whisper room DOES contain metal music stand draped with towel, and an undraped wooden table on which is located a computer keyboard and mouse. Thanks to you, I was able to load the ACX plug-in. When I tested the raw clip with the ACX check, with recording level at one-quarter, peak level was -20.0 dB (passes), RMS failed at -39.1 dB, and noise floor passed at -61.6 dB. I did several clips at recording level set at midpoint, but the RMS AND floor test failed. And at high recording levels, the RMS passed but the noise floor really failed. So I would sincerely appreciate your analysis and any suggestions.
That explains why it’s quiet : the maximum peak value permitted by ACX -3dB , not “-20.0 dB”.
The RMS value permitted by ACX is in the range -18dB to -23dB , (i.e. about -20dB).
Once you turn up your recording level so you have peak volume values of about -6dB that should solve a lot of your problems , ( the signal-to-noise ratio should become better ).
I would. Reflections from the table can affect the timber of your voice. The drape also suppresses table thumping and script shuffling noises.
In addition to suspending the microphone, do you have a “spider” vibration isolator and pop/blast filter?
This is a fully outfitted “studio.”
There’s a moving blanket on the table and that white spider thing prevents vibrations from the floor from getting into the microphone.
The black tennis racket suppresses breath sounds and “P” pops.
This is a little Hollywood because you would never place the microphone that low and that arrangement only works with quiet Macs. That design is after an actual instructional video on the ACX web page. They were using a soundproof room somebody built in their house. Also, no studio I ever saw had the cables arranged that neatly.
Starbucks is optional. Coffee Bean and Tea Leaf works , too.
The regular Yeti is a USB microphone. How long is the connecting cable? Audio doesn’t like going down long USB cables. If you have troubles with clicking, popping, snapping or small sections of sound actually missing, that cable is the first place to look.
Missed a step. We publish Audiobook Mastering 4 which is basically three tools that push your presentation into ACX technical compliance—except for noise. There is another whole series of postings what to do if you fail noise.
Also note after all this fuss, you only have achieved technical compliance. An ACX submission still has to pass Human Quality Control where your submission may die if can’t read.
There is also one slightly more exotic failure. USB microphones may have a failure I’ve called The Yeti Curse (turn the volume up a bit).
That can prevent you from passing ACX Noise and it can kill Quality Control Acceptance. It doesn’t submit to Noise Reduction well and it’s difficult to get rid of.
Also note that plain ordinary Yeti and Professional Yeti are the same thing. Yeti Pro® is a completely different microphone (at twice the cost) and as far as we can tell, doesn’t have that noise problem.
Thanks very much, Coz! I’ll start with the chair and table drape, then rerun the ACX test. (I’ll also substitute Folgers Black Silk for the Starbucks…)
coz, this was done with new chair, draped table, boom mic “spider”, fully-draped stand. Still doesn’t pass ACX noise floor. Should I go to pedestal mic on table? Might use a shorter USB cable… Anyway I sure would appreciate your comments.
The catch is I had to design a custom filter to get rid of what I’m assuming is computer fan noise. You have a background hum or whirr at 81Hz musical pitch. It gets reduced in the Low Rolloff equalization step, but not enough to pass. 81Hz is not a regular hum sound such as 60Hz in the US and 50Hz in the UK. It’s such an odd pitch that it almost has to be computer fan noise as computers don’t have to follow the power system like an air conditioner or other room vent fans.
Even in the studio, can you tell if the computer is on just by listening?
You can do the rest of the work just by adding the custom filter before you apply Mastering 4. It shouldn’t affect your voice.
Effect > Notch Filter > 81Hz, Q 6.
Don’t change the computer. If that’s what it is, you are now married to that computer’s fan noise.
So that’s the technical considerations. That leaves theater. This is where I take one step back and you get to decide if you like the voice quality and everything about the sound. I think it’s fine as it is, but there are very few voices where somebody can’t find some way to change it. We should remember that your mastering process is already 25% longer than regular mastering (four steps instead of three) and you will need to remember to apply all the corrections to each chapter. Anything else you do is going to be in addition to all that.
There’s a lot of infrasound in that , which no-one can hear but apparently is measured by the ACX tool.
If you remove those inaudible frequencies with this rumble-filter equalization, it passes ACX noise-floor test …
I’ll definitely do the notch filter. But note that the computer is totally outside the whisper room, just leaving a solid-state keyboard inside on the table.
But note that the computer is totally outside the whisper room, just leaving a solid-state keyboard inside on the table.
How are you getting ventilation in the room? Mini whisper fans? Can you hear them change if you unplugged them?
We had a conference room at work which was soundproofed. I sent a number of sound performances through that room. Yes, it had padded walls, etc, but it also had padded air ducts and I couldn’t tell when the air conditioner went on.
If someone unplugged the computer with you in the room, could you hear it wind down? That whir tone behind your voice is coming from somewhere and it has all the signatures of a computer fan. If it walks like a duck…
There is an insane test you can do with a long microphone cable. Look for the hum like that guy at the beach looking for Rolodex Watches in the sand with a metal detector.
It’s almost certainly a directional microphone so you aim it all over while wearing headphones and set Audacity for live monitor. The whir will get louder as you approach the noisemaker. That’s how I found my music bass cabinet didn’t go off when I turned it off and it sat there making a low volume hum that drove me nuts for a year.
Tried notch filter on my computer with Audacity 2.3.0 (see attached waveform). Failed ACX noise floor -55.5dB.
Fully mastered and Eq only pass (see attached) (-64.0 and 61.3, respectively).
Please listen to the audio. I’d love to know whether sound/theater quality of Eq only and fully mastered are OK. Also, could you suggest why my notch filter didn’t pass? Also should I just master using Eq to minimize actions that might fail ACX sound/theater quality?
Also, could you suggest why my notch filter didn’t pass?
The notch filter should be used in addition to the other processing. I only notches-out 82Hz noise.
Also should I just master using Eq to minimize actions that might fail ACX sound/theater quality?
The default low-frequency roll-off is just to reduce low-frequency noise. And, since there is no deep bass in voice recordings the low-frequency roll-off doesn’t affect the (good) sound.
Additional EQ can be used (sparingly) to as a corrective-effect to correct for microphone frequency response variations (i.e. if you microphone sounds “dull”, etc.) or as an artistic or special-effect to slightly-change the character of your voice.
