There’s no set procedure… Live radio announcers use almost no processing except there is always a limiter because it’s illegal to over-modulate.
Most books & commercials will have compression/limiting. Limiting is a kind of fast dynamic compression that “pushes-down” the peaks. Then make-up gain is used to bring-up the overall loudness/intensity.
Ideally you won’t need noise reduction but if you want to approach “professional quality” without a soundproof studio you’ll usually need it. Noise is usually the biggest problem and if the noise is bad, “the cure can be worse than the disease”.
There WAS an audiobook procedure but they killed the wiki… It was a high-pass filter to kill any noise below the voice frequencies (maybe high-passed at 80Hz but I don’t remember). Then Loudness Normalization was used to bring the volume up to about -20dB RMS. And finally limiting… The -20dB RMS adjustment usually pushes the peaks into (potential) clipping (distortion) and limiting can fix that (if done before you export).
For ACX audiobooks the requirement is for peaks below -3dB so the recommended limiter setting was -3.5dB. But for regular voice-over you can probably hit 0dB (the “digital maximum”).
If you use regular (peak) normalization do that last.
Or limiting (which also affects peaks) can be done last.
I would just experiment… AM radio goes from about 50Hz to 5kHz… But some radios will limit it more. When you limit the high frequencies you reduce the “T” and “S” and that hurts intelligibility.
Some distortion may also help with that old-time sound. the Limiter has a setting for Hard Clipping. and there are other Distortion effects. Clipping is the most common kind of distortion… You get clipping when it you try to get 11 Watts from a 10W amplifier, etc. It can also add “grit” to voice.
AM radio usually also has noise in the background (hiss and/or hum, etc.). You can use the Generate → Noise tool to generate a couple kinds of noise and mix that in, if you like the effect.
The Audiobook Mastering software is still available and the process can be regenerated if needed.
However, just to start from closer to the beginning. Do you have a quiet, echo-free room to record in? If you do, then any kind of microphone and performance process can be made to work. If you don’t, you will spend an impressive amount of time patching the errors and distortion—if even you can.
From where I’m sitting right now, I have a wall clock ticking, the refrigerator pumps and cooling fans kicking on and off. There’s the Metrobus every half hour or so. and I can usually tell when the traffic light changes up the street because a knot of car traffic goes by.
I have one table lamp which “sings” at a high pitch when it’s dimmed, and one of my office machine’s hard drives is an older one and I can hear the actuators clicking back and forth.
Can you hear your computer’s cooling fans? You will need to be able to see the Audacity sound level meters while you’re performing and a noisy computer is really bad news.
Is this your first forum post? The forum software has a thing that it will not let a New User post a voice test. So we’re flying blind, or deaf in this case. Do you have any way to post a short sound sample and then give us the address?
Have you plugged in the Rhode and tried it? Recording on the computer can be really painful with other software packages trying to “help you” with your recording. Do you use Skype, Zoom, Meetings or other communications packages?
My stand-alone recorders all record “flat” to avoid imprinting a personality on the files. It’s good to know the “warm” settings.
I had access to a super-duper microphone once. I liked what it did to my voice. I did the crunching newspaper tone test. It was anything but flat. It was an eye-opener. Ear-opener?
I know I can get the “intimate” effect by recording with the microphone opposite my cheek (to avoid P-Popping) and closer rather than straight-on.
First of all I specify that I use a translator, if there are grammar errors it is normal. ^^
I send you my recording:
I’m in a 10m² room with no extraneous noise, apart from the computer fan, but I put myself at a good distance to minimize the noise.
I send you too the recording with the following effects:
Noise: 6/6/6
Compressor: default except ratio: 4:1 instead of 2:1 and attack time of 0.10 secs
EQ: boosted bass + treble boost
Normalizer: -1db
Limiter: default (hard limit)
I have a question about the normalize effect.
As I record myself between -18 and -12db, when I normalize everything goes to -1db at the maximum, which makes the volume arrive in the yellow/red zone and I find that it is less clean. Whereas if I normalize -12db I have the impression that it gives a better result. Yet many people advise to normalize to -1 or 0db… why?
The fan noise is loud enough to make a professional ($$$) quality recording impossible : the noise-reduction required to remove the fan conspicuously damages the speech …
Set limiter to “Soft limiter”. (not hard).
Audacity’s compressor is not suitable for processing speech:
because it’s release-time cannot be set shorter than 1 second.
The free version of TDR’s Kotelnikov compressor plugin is worth having.
There is a noise-gate plugin everyone should have: Couture.
The free version is sufficient. (Made in France ).
I applied the simple Audiobook Mastering Macro and it passes all three ACX sound specifications.
The background noise “passes…” but is a little too loud. You should hit noise at least -65dB or quieter. I applied the noise reduction of the beast (6,6,6).
If you can make the computer fan noise go away, I bet you won’t need the noise reduction at all. There’s a rule here: If you can tell the computer is on by listening, it’s too loud.
And that is the mastered and submittable MP3 sound file. There are still some theatrical problems, however. Lip smacks, gasping for air, etc. That’s exactly what killed my announcing career.
The sound is OK, but you sound like you’re tired and really want to be somewhere else. Now try one where you’re trying to sell me something. You can go longer than ten seconds. You can take air, but don’t smack your lips.
I recently bought some groceries and there was an announcement for sales help. I repeated the announcement in perfect tones. The checker looked at me with big eyes, but the bagger collapsed laughing.
You can experiment with other patches, effects, and filters if you want, but I think you’re good to go just like that. I never got good covering up my lip smacks. It’s a tricky editing technique. You do that before any mastering or other processing.
Your posted WAV file is slightly low volume.
The strongest tips of the blue waves should come in around 0.5 or slightly quieter. That’s -6dB to -9dB on the bouncing sound meter.
Mastering is a text file. This is a picture of the exact file.
It’s way more awkward than you think because computers keep trying to “help you.” Most cover up the .txt filename extension. So you’re never quite sure what kind of file you have without making the computer explicitly tell you.
Then, it is a text file so the computer will try to treat it like an English word document rather than a series of precision effects instructions which is what it really is. @#$%.
There are instructions how to install it and apply it.
Also highly recommended is ACX-Check. That checks the technical quality of your work a lot like the ACX Audiolab tools do.
Oh, and to drag this home. Most times, if you can achieve ACX Compliance, you can post your work almost anywhere else, maybe with minor volume adjustments.
Also remember, if you fall in love with a fist-full of effects, corrections, filters, and adjustments, you’ll need to apply them all, every time, to produce work. Some, like Noise Reduction, can’t be automated.
I just noticed the makers of Couture have a new plugin called Lens
Multi*band compression & equalizer, (real-time effects).
The free version is sufficient.
(Allegedly compatible with Windows, Mac & Linux)
: the noise-reduction required to remove the fan conspicuously damages the speech …
].