3

I know this is probably been said, and I know there is a preset on the effects panel in Adobe Audition. Though using a preset feels like cheating and samples will sound the same. I want to know is there a way to make it a little more convincing? In the sample below there is mouth breathing, is this done during recording or can it be done after with distortion? and how?

around 41:40 or anywhere with call ins.

|improve this question
2

some light phone handling added in along with the correct EQ and a speaker simulator will usually get you what you need. Or a plugin like Speakerphone will let you get to a good place very quickly also.

|improve this answer
  • Yes, Speakerphone is the gold standard for this, it will nail exactly that — and many other types of simulated spaces effects. It's incredibly versatile and fun, it has impulse responses for various types of telephones and a mic degrade option to further grit things up. If you find it too expensive, look into Plogue Chipcrusher and McDSP Futzbox. – TORLEY Sep 4 '15 at 22:27
1

Put a high pass filter at 300 Hz, a low pass at 3300 Hz, and compress the heck out of it (compression will help with the breathing noises). You might want to add a little distortion to replicate the carbon microphones that at least used to be common in telephones. If you have a chance to record the audio, put the mic as close to the talent's mouth as a telephone mic would be to better replicate capture of breathing and mouth noises. And I would just use a cheap dynamic mic for it.

|improve this answer
  • That's awesome, thank you for quick response. – GentlemanSquid Sep 4 '15 at 20:49
  • Why these values 300Hz and 3300Hz ? – Abdul Al Hazred Sep 9 '15 at 15:01
  • 2
    @AbdulAlHazred Long ago the phone company (in North America at least) decided that it wanted to design for a 3 kHz bandwidth and wanted to get the most important frequencies for human speach. Most voices don't go below 300 Hz in general so they started there and went up 3 kHz to 3.3 kHz. One issue (in terms of phone calls) is that sibilants and fricatives have important information at 5 kHz and above, so on the phone it can be hard to tell an "f" from an "s". See: en.wikipedia.org/wiki/Plain_old_telephone_service – Todd Wilcox Sep 9 '15 at 15:40
  • 1
    With replicating this sort of degraded sound quality, a bitcrusher is useful to bring the sample rate down to 8kHz, which, if I remember correctly, is the sample rate of a standard telephone signal. Without doing research, I'm not sure about the bit depth. – Marc W Sep 13 '15 at 4:37
  • I think an ISDN B channel actually carries 8 kHz x 7 bit words for a total bandwidth of 56 kbps (the other 8 kbps are used for control data), hence the maximum modem speed. You could replicate that with a 7 kHz bit crusher if that is available, or 8 kHz if that's the closest available setting. Good idea! – Todd Wilcox Sep 13 '15 at 6:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.