Exported MP3 file is removing vocals on PA system

Windows 10, ver 2.1.0

MP3 That is exported from audacity with a music and a vocal track is stripping out the vocals only when played on PA announcing system. The MP3 plays successfully with both music and voice recording on both Windows 10, ipod and android phone using headphones or speakers. Tried exporting to a wav format but same problem exists. Any suggestions or thoughts appreciated on what the problem may be.

Oddly, this isn’t unusual. The PA system is using a mono mix of the show and all the other methods of listening are in stereo. Somehow, you managed to produce the vocal part of the show out of phase between the left and right portions. As long as you never mix left and right together, the show will appear to be OK, although the voice may be a little odd sounding.

You can test for this in Audacity by mixing left and right temporarily. Tracks > Stereo to Mono. I bet the voice vanishes.

No doubt you still have the vocals and the music and other sound on separate stereo tracks. Split the vocal into independent tracks, reverse the phase of one and then recombine everything. If you only have the final mix, you’re dead.

Let us know.

How you got there is worth investigating, too. How are you recording your voice—in detail with part numbers.


Thanks. Changed the Tracks>stereo to mon for the mp3. Couldn’t reproduce the problem. Vocals were there.
Only have one vocal track. Can you explain how to reverse the phase and recombine? New at this, first time using this app.

Also, I have a copy of the music but looks like I lost the voice track when I did the export. I am going to redo the vocal track. Any tips when doing this again so same problem doesn’t occur when exporting to mp3?

OK, that was the easy, low-hanging fruit. Now it’s harder. Describe exactly what you’re doing. Where is the music coming from and how, exactly, are you recording your voice? Model numbers or web pages. Pretend I want to buy the same stuff you have.

I lost the voice track when I did the export.

It’s bad practice to do editing, production, filtering and effects on original work. If anything goes wrong, you get what happened to you.

When you’re done recording your voice and before you do anything else, select the whole recorded voice track by clicking just above the left-hand MUTE button.

File > Export Selected: WAV (Microsoft) 16-bit. Call it something like Raw-Voice. If you need dates, use the ISO form. Today is 2016-11-06. That’s your backup voice. I assume you got the music from somewhere where you can get it again. The goal is to be able to reconstruct your whole show from high quality parts and files without re-recording if something bad happens.

The truly obsessive/paranoid [raising hand] will save copies of all original work on an external thumb drive. You probably don’t need to do that unless somebody is writing you checks.


I think the problem is the balanced (3-wire) connection to the mixer. I assume it has XLR or TRS connectors? And you’ve connected a stereo output to one of the mixer’s mono inputs? Those are differential inputs and if you use the “wrong” adapter the left & right channels are getting subtracted.

The best solution would be to connect the left & right channels to two different mixer inputs and allow the mixer to mix the left & right channels normally.

If you don’t have an extra mixer input channel, make a mono mix and get a cable/adapter that allows you to connect the left or right channel (with the opposite side unconnected).

I think the problem is the balanced (3-wire) connection to the mixer.

The poster hasn’t told us he has a mixer.
That’s why I asked to describe how the voice is being recorded. You can also get into trouble like this by badly wiring an XLR microphone into a soundcard.


The music is a downloaded song. Voice was recorded on computer using computer microphone in audacity. Still have music track but did lose voice recording. Will re-record voice and save separately so have a copy. Any suggestions on how to properly export after this is done so I don’t recreate the same problem?

using computer microphone in audacity

Again, we don’t know what caused the original problem and it’s dangerous to continue until we figure it out, so you’ll have to get closer. Pretend I want to buy your microphone. Do you mean the microphone inside your laptop?

We both immediately went down that XLR microphone pathway because this sound damage problem can happen to new users with higher-end microphones and sound systems.

But it’s not so common with a USB microphone or whatever you have.


I’m using the microphone in the computer. i have started over and made the voice and music on the left side speaker to see if that will make a differences.

Rather than that, you can make the system record in Mono (one blue wave) instead of stereo. We’re almost all united in thinking it has to be an odd stereo problem that the PA system can’t handle.

Mono should come out of both speakers even though it’s a single blue wave and not two. It should say MONO in the INFO panel to the left of the track.

My laptop will record internal microphone either stereo or mono and will follow my settings.

You can collapse the music to mono and do the whole show in mono. Select the music track by clicking just above MUTE. Track > Stereo Track to Mono.

If you’re in love with the violins on the left and French horns on the right, that’s going to stop when you go MONO. Everything comes out of both speakers.