newbie seeks recommendations

[kozikowski]

Did I ask you to intentionally overload the microphone by yelling into it? Never blow into a microphone, but yelling is allowed.

Do one of the 10 second sound tests but instead of reading a passage from the milk carton in your normal voice, say "Loudness test, one two three, ONE TWO THREE" and keep getting louder and closer to the microphone. Warn the neighbors. I want you to intentionally overload the system.

Stop, Export as WAV and post it.

There are ways to test if your reading is going to be damaged before ACX sees it, but I need this overload test first.

Thanks.

Koz

[kozikowski]

Any of those stingers work for you?

Koz

[Lorain]

Thanks! Good to know it's not something I --maybe-- have to worry about as long as I keep my voice down.

I thought I had left in the right amount of room noise at the beginning and end, but apparently I must not of. I don't understand that because I was very cognizant of those requirements.

I tried the stingers but none really seemed to work for me. I ended up using Audacity's Risset drum. Thank you for your efforts on that.

I'll send in another sample, see how it goes.

I have however learned one very valuable lesson for all my future writing: never write a novel and give the main character a name that starts with a "P"! I just thank goodness I didn't give her a name that starts with the "Sh" sound!

Thanks again, especially about warning me off of the MP3. I'm not sure they accept WAV, but I will try.

[kozikowski]

I'm not sure they accept WAV

They don't. They require MP3 like in two or three posts ago.

Your work should all be in perfect quality WAV. Open the Edit Master WAV in Audacity and Export > Export as MP3 for them.

You can't edit or manage an MP3 without adding sound damage. They put up with the damage as a corporate decision.

Koz

[Lorain]

Oh sorry, I thought you meant to export to ACX in WAV which I didn't think they'd take. As soon as they give me a response to my second sample, I'm sure I will be back asking for your help.

[kozikowski]

Posting their response was a big help. Some performers just tell us "they didn't like my work," and we get to go round and round figuring out what exactly they didn't like.

I won't mind knowing whether your microphone technique or your mastering created the distortion so we can tell other people.

We know what ideal announcing volume is. The bouncing sound meter should, on occasion, just start turning yellow, around -10dB to -6dB.



This is the down side of putting your computer screen where you can't see it.

Koz

[Lorain]

I checked. Yeah, I was pretty much yelling in the part they objected to. I've attached the part of the file they objected to (had to shorten the sample to fit here). It's the part where the character is yelling "POLONIA! POLONIA! Where are you?". I will have to tone it down I guess. Unless there is a Filter on Audacity that will do that for me automatically?

[kozikowski]

I was pretty much yelling in the part they objected to

Attached, something in your system ran out of moxie (technical term). See all the blue tips in the "Polonia" monolog line up?

overload.jpg (57.75 KiB) Viewed 171 times

Many notes there.

-- Audacity doesn't apply filters, corrections or effects during recording.

-- Overload like that is called "clipping" because the blue waves look like somebody came by and clipped off the tops and bottoms. Chances are good it's permanent and fatal. The digital system stopped following your voice. There is no "clean" voice to rescue. It's a do-over.

This is where you ask about re-recording just the "Polonia" part. It's possible, but by the time you get good enough with the editing tools, you could have read the whole chapter again. Sometimes, reading one or two sentences around the gaff works. Then you open up the perfect quality WAV you made of the original recording and lay the perfect quality WAV patch into it. Nobody would object if you read the revision several times and picked the one that sounded best.

-- Some performers back away slightly when they yell so the stress of the voice is apparent, but the voice doesn't actually get louder. Even better if you can announce the stress without backing up or getting louder. Live headphones are terrifically important here.

-- There may be semi-automatic ways to tell just after performing that you have a damaged track, but I need that overload test to find out.

Koz

[Lorain]

I am trying to turn on Audacity's "Show Clipping" feature. I click View > Show Clipping, but nothing happens, i.e. even with the box checked I don't get the red lines showing the clipping. Is there something else I'm supposed to do?

Also, I'm trying to use Audacity's Clip Fixing feature. Do you by any chance know what the loudest Reduce amp number for ACX is, that it will accept?

[kozikowski]

loudest Threshold for Clipping, and, Reduce amp numbers are, that ACX will accept?

ACX doesn't care about clipping. They care about audible sound damage—that gritty, crunchy sound that generally happens when you abuse the microphone system by performing too loud.

Effect > Clip Fix is intended to hide obvious damage of a very brief sound—small fraction of a word—that's too loud. And it doesn't fix anything. It guesses what the original sound might have been based on the surrounding sounds and tries to make everything sorta match. It works in the subjunctive if you remember your English lessons.

It won't fix a whole word much less a whole sentence.

View > Show Clipping is intended to reveal, with its little red lines, any sound whose blue waves fill the space top to bottom. You would think this would be obvious, but it's not when you're looking at a complete chapter performance or even longer.

It gets more magic when you reduce the volume of the performance after the damage. The crunchy damage doesn't go away, it just gets quieter, and the performance appears to pass Show Clipping.

Koz