From phone effect

GoldWave general discussions and community help
Post Reply
Nanashi
Posts: 2
Joined: Wed Oct 11, 2006 7:36 pm

From phone effect

Post by Nanashi »

I'v got a sound file of one of my voice actors talking that I would like to convert to sound like it was recored from the other end of a phone (like a caller on a talk radio show) and I just can't get a chain effect put together that does it convincingly. I start off with the EQ band and turn most of them down, to cut out the base and trebble, and then I mess around with making it a little less clear, but its just not turning out.

Any suggestions?
DougDbug
Posts: 2172
Joined: Wed Feb 16, 2005 3:33 pm
Location: Silicon Valley

Post by DougDbug »

Maybe some compression and a little distortion. I wrote the following in response to a question about simulating a 2-way radio:
DougDbug wrote:On a real radio, the poor sound quality is due to four things:

-Limited frequency range
-Noise
-Distortion
-Compression

The distortion and compression are related, as they are caused by overloading the circuitry with too-strong of a signal. A bit of this overloading can actually increase inteligibility by drowning-out noise.

So, you might try some compression - Take-out the level variations, so that all parts of the speech are very loud. There isn't a preset for this. Maybe start with Boost Low Levels, and set the threshold at or near 0dB so that everything gets boosted except the loudest peaks. (You can reduce the overall volume when/if you get the sound you want.)

You might even want to force some clipping (which is an extreme form of distorted-compression). GoldWave tries to prevent clipping, so what you need to do is boost the volume level so that it exceeds 0dB. Then save the file. When you re-open the file it will be clipped to 0dB. For a starting-point, Maximize the file first, and then increase the level by +6dB. Make sure to save your original file with a different name, because once a file is clipped, you can't un-do it. (And, you will need to experiment with different amounts of clipping.)
Kummel
Posts: 141
Joined: Sat Sep 23, 2006 7:10 pm

Post by Kummel »

I suggest you to try, and adjust if needed:

highpass 300 Hz
clip the voice (with the dynamics effect)
resample at a lower sample rate (8000?)
resample back to the normal rate (44100 or 48000 Hz)
amplify treble to make the resampling noise more audible.

Edit: Sorry, didn't see Dougbug's post. :(
:D
Nanashi
Posts: 2
Joined: Wed Oct 11, 2006 7:36 pm

Thanks

Post by Nanashi »

Thank you both, I used a little bit from each post and was able to produce a much more convincing effect. I'll try and post a link to the video once it is done so you can see what I'm talking about :)
Post Reply