I’m new to the audacity forums and I’m not sure where is the best place to make this post. If your a moderator feel free to move to this where it belongs…
The main thing I use Audacity for is fixing waveforms. I don’t have enough knowledge to be mixing music, but sometimes I use Audacity for correcting or customizing podcasts and sermons.
I ran into an issue today that I have never experienced before, and a part of me hopes I will never experience again. I downloaded a sermon from a church website. When I played it back in stereo head phones everything sounded good. However I plugged in one of those iHome mono speakers and everything sounded awful and demonic. lol I tried another recording and it sounded fine through the speaker. I know there is nothing wrong with the speaker. But for some reason the sermon didn’t sound right in that speaker. After doing some thinking… I came up with a theory. My theory was one of the channels, (Left or Right) I didn’t know which, might be inverted. I tested my theory out by using Audacity to invert one of the channels. When I played it back through the same speaker everything sound good. I wonder two things. Who, What, When, Where, Why, How did this happen? and Second when your looking at two waveforms, is there a way or telling if one is inverted?
If one track is inverted, you can usually see it if you zoom in close.
Note that in a stereo recording that has a wide stereo sound stage, it may not look as clear as in the above example - there may be some moments when one channel appears to be inverted (and at that point in the track, is inverted), but what you would be looking for is an overall pattern. The waveforms in the channels should, overall, look to be the same way up.
Hard to say without knowing how the recording was made.
One way:
If there were two microphones recorded via a mixing desk, the mixer may have an “invert” button for each channel and one could be pressed and not the other.
By the way, it does not matter which track is inverted. If both tracks are inverted, no problem. The important thing is that the “phase” is the same in both tracks.
A negative value for correlation indicates strong cancelling when mixed down to mono. -100 % is the maximum and means that both channels are the same with one being inverted.
Lately, I see a lot of audio on YouTube that is inverted in that fashion. I wonder what the purpose can be of it.
It sounds awfully fuzzy. Downloading a low-resolution video of it gives you one audio channel which is accordingly empty.
There isn’t really a “right side up”. Sound as we hear it is “air vibrations”. The waveform in Audacity represents those vibrations as a wiggly line. Depending on which way round the microphone is pointing, which way round it is wired, how far from the sound source is from the microphone, etc. the waveform could be either way up and still be “correct”. The important issue is the “phase” relative to other waveforms.
There is a convention for microphones. A positive air pressure at the microphone should result in a positive-going waveform. I did an experiment I haven’t posted yet where I struck a piece of paper in front of a microphone such that the air pressure between the paper and the microphone went up. Sure enough, the waveform at the output of the mixing desk went positive.
That’s not an unconditional test because the performance could have been flipped several times in the chain, but that’s not likely.
There is waveform sense in the wild. Some voices have a pronounced asymmetry. I can recognize some broadcast voices strictly from the waveform.
“I can’t tell those two voices, but that one is probably Molly.”
