for a project I have to record lots of conversations via smartphone. One big problem at the moment is cross-talk.
For Video production I use my Zoom H5 recorder. Usually this works fine but for this project we need to have lots of cross-talk happening timed perfectly. So it is hard to record the audio as one stereo file as both speakers are cross-talking.
In the Scenario of 1:1 phone conversation:
My plan was / is to have the Headset and Phone connected to the H5 recorder.
The recorder should record the Stereo speaker signal coming from the phone in channel 1 and in channel 2 my mic input.
For an Android Phone i used the CTIA standard of the plug. I connected two cables to the the corresponding cable pairs. Channel 1 of the XLR Plug is connected to Ground, left and right speaker. I then connected the channel 2 XLR Plug (wires: ground, hot, cold) to the mic line of the headset (wires Ground to ground, mic to mic line).
First problem is that if i only listen to the mic line, i can still hear the speakers. It must be some bleed into the other lines from the ground line?
But the main issue is that I can not seperate the lines this way. the mic is just dead when I connect it this way. It does work perfectly without the bridge.
What am I doing wrong?
Could it be that Android is recognizing difference impedance and switches the layout?
What could I do differently?
Any help is greatly appreciated. I have been trying to solve this for two days.
What you've written makes no sense at all. XLR audio connectors don't do stereo. Phone conversations aren't stereo. Connecting a mic into two inputs at once generally doesn't work either. And in an audio context crosstalk is when the signal from one channel is picked up by another channel which doesn't appear be what you're talking about.
Have you checked the pinout for an XLR connector and for a headphones jack?
Yes, audio is transmitted of course only in mono. The output is split / combined onto the stereo channel L/R. That is what i meant with stereo. The mic output from the headset is connected to the XLR channel 1 and the stereo (L/R) is connected to the XLR channel 2 (bridged). You may also just connect only the left or right channel.
Crosstalk is meant in this context as both speakers talking at the same time. I know, it is misleading in each context. Does mean a different thing in audio then in video. In Editing different again. But in this case It is meant when to of my interviewers talk at the same time. I want to split my input from the feed.
I'll add a quick circuit plan. (to the original post)
I was wondering if it is a problem to connect the XLR channels to the shared ground?
Do I maybe need to use plug-in power?....
OR can it be that the signal control from Android is preventing this to work? --> as like shorting the L/R channel is like a long button press etc.
Hi Paul, the thing is that the counter part (speaker on the receiving phone) can not hear me.
I am trying to figure out if i made something on the signal level, super basic wrong.
Also there is a constant noise wave in the background. I scoped it and it is 100mhz @ a couple of mV.
Avion32:
First problem is that if i only listen to the mic line, i can still hear the speakers. It must be some bleed into the other lines from the ground line?
Yes, that is unavoidable if the ground line is shared as the resistance of the ground generates voltage
due to the speaker currents, which is seen straight across the microphone input.
I'd suggest using a separate microphone to record, don't try to use the headset mic, you'll probably
get way better quality too as headset mics can be pretty mediocre.
The other approach is digital processing to reduce the crosstalk - this has a good chance of working
well as the leakage is electronic rather than acoustic.
You keep saying XLR but have pictured a jack. If you're using XLR connectors there's a good chance you're using the wrong pins.
Mics are simply not designed to feed two separate devices. You will need to use a splitter to feed each one. Or as MarkT says, use a second mic instead of the cheap nasty one in the headset.
Hey
thanks for the answers already. @MarkT, that is what I assume of course. I can filter it out quite easily afterwards though.
BUT the weird thing is that the mic is not working anymore when i set it up this way. Would love to understand what is actually happening. Like is the mic internally maybe preventing smth. as it then has too much resistance or the phone?
Because electronically I can see the signal from the mic.
@Chopsuwe, i was hoping you don't pick up on that YES you are right, in the circuit plan are 3.5mm jacks. For the lack of a better item i used that one but labelled and connected it correctly. I could not find quickly a xlr plug. (why ever).
In the pictures though you can see that these are XLR Male plugs going into the Zoom recorder.
For now at the moment I am doing it like you guys already recommended. I use two seperate mics. It is just cumbersome but I think you are right, trying to make or break this for what can only be a bad solution is probably not worth it. Rather connect a second lavallier to the mic and just simple lay it along side the cable.
I don't doubt you have it plugged in, I can see that. I asked for a very good reason which I'm not going to repeat for a third time. Why bother asking for help if you're not going to listen?
@Chopsuwe I really don't perceive properly why you feel threatened or behave this agressively.
As I said and illustrated in each response to you, I actually did listen to every word.
First of all, they are XLR Plugs and they do have two leads that can be used in a different standard dual inputs.
I did reply to both of your rather unpolite responses with detail and took the effort to quickly draw you a circuit layout to illustrate that the connection is sound. YES I did not find an xlr plug and used a 3.5mm cti jack instead while explaining that this is not the representation.
Also and furthermore I did change the layout, corresponding to your first and second post, to supply the inputs with seperate power each individually so that they do not use a common ground.
Lastly I replied in a polite manner to all your unpolite and rather flat out insulting postings.
Henceforth I agreed with Mark that it must be rather make no point to try to achieve this with an inferiour design.
It does work though. I changed the resistors and applied a phantom power to the inputs. Due to the common ground the noise is of course in the background but I filtered that out with a simple a.i. model tought to listen to the frequency modulation.
Hence I wonder what is your point? Are you trying to insult me or bring this in any way up? I found a solid solution to my problem, received help from mark who was able with one post to detail more information then you did with multiple posts. Now that I read your stuff I rather have the feeling you write here just to gain visibility and prove a point to yourself instead of helping others. Always remember that one may not be an exptert in one field but may very much be an expert in many other domains.
YES you are right, in the circuit plan are 3.5mm jacks. For the lack of a better item i used that one but labelled and connected it correctly. I could not find quickly a xlr plug. (why ever).