Help on interviewing someone over Skype on podcast!

Okay, I figured out how to do it with sound flower, however I haven't done it over Skype yet. Obviously I have to mute the track in Logic or I'll hear the doubling effect when playing clips. So, I want to create an aggregate device correct? Then he can hear what's on my computer and my mic? Since this will allow him to hear what's playing on my computer won't it make him hear some latency in his voice?
 
Do I need to choose 16CH instead of 2 CH in order to do that?

Oh, right enough..yes.
You'd need to use SF1+2 as your 'to skype' bus and a different pair (3+4?) as your 'from skype' bus.
I have to admit I hadn't processed that. I take it Skype and other apps will let you pick specific channels out of all the available SF I/O?
If not you'd have to find another way to get the Skype call output into logic but that shouldn't be too hard, even if you have to use a hardware loop it'd be worth it.

You'd still be able to make great use of SF as an output bus for the interviewee regardless, whilst being able to multitrack record and process everything as it happens.

Okay, I figured out how to do it with sound flower, however I haven't done it over Skype yet. Obviously I have to mute the track in Logic or I'll hear the doubling effect when playing clips.

Um, I'm not sure you have it figured out!
Everything goes to logic. You distribute what you want to yourself (main output) and the other guy (SF 1+2) from logic.
There should be no double paths. There may be minimal latency but no one wants to listen to themselves, right?


Now, how do I do that? I don't see those sort of options?
What? There are no options. SF is just like a logic bus except other apps can see it.
Your options are all in logic.


Admittedly it's a pretty intimidating set up if you're not familiar, but it's all in the name of discreet recording and full control of everything.
If that's overkill for your purposes, certainly go with something simpler. :)
 
Last edited:
Well, I'm slightly confused. Okay, in sound flower, on the 2ch mode. I keep those on none and make logic input sound flower 2ch, and it does record my computer audio on there, as well as Skype. I can also record my mic separate as I made a combined device.

However, I don't see how I can choose what I send to the Skype guest? If I choose sound flower as input on Skype, he will hear everything that happens on my computer.

I can't really figure out how the routing works. If I put it on anything other than none, it just gives me a double effect.
 
Well, I'm slightly confused. Okay, in sound flower, on the 2ch mode. I keep those on none and make logic input sound flower 2ch, and it does record my computer audio on there, as well as Skype. I can also record my mic separate as I made a combined device.

However, I don't see how I can choose what I send to the Skype guest? If I choose sound flower as input on Skype, he will hear everything that happens on my computer.

I can't really figure out how the routing works. If I put it on anything other than none, it just gives me a double effect.

I think I see what you're doing. You've set the system output to SF 1+2?
There's no need to do this.
You're simply using SoundFlower as a bus, just like you would in logic, except it has the advantage of being seen outside logic.



To simplify the concept, imagine you're having a call and you want to play the person some music, but not through the mic.
Instead of setting the Skype input to mic, set it to sound flower 1+2.

Now go into logic (with aggregate device).
Get your mic input into logic and set the track output to whatever numbers correspond to SF 1+2. Hit record arm. The guy can hear you again.

Now why not set up a new track with some music on it and set its output to SF1+2. Hit record arm.
The guy can hear that too. :)

To extend that, put a compressor and eq on your voice. He can hear that too.



Now why not take your second mac and run it into line inputs 3+4 (or whatever) of your interface and set a track up for that in logic.
Set track output as above and he can hear that too!

You can now record all the live sources from your side discretely into logic whilst sending it to the guy 'mixed'.
See?

You can create a submit in logic to feed whatever you need to hear to your main outputs as normal.
Does that make sense?


The only thing you need to do in addition this this is use two different soundcloud channels to get his voice into logic, for recording purposes.
You'll just output that to main 1+2 in logic so that you can hear him.

*This is the bit I hadn't figured out initially. If Skype doesn't offer you SF3+4 in the options, then you'll need to do something else.
You could set Skype out to hardware line outs 5+6, for example, and cable route those back into line in 5+6 for logic to see it, then just record as normal.
Of course, don't sent it back to him via SF 1+2. There's no point.


Your OS system device is largely irrelevant, but it may as well stay as hardware 1+2 as normal, or whatever will feed your headphones.
Setting it to SF or whatever will mess things up.
 
Out of interest, did you get Skype to show the separate sound flower channels? 3+4 rather than 1+2, for example?
If this can be achieved with the focusrite software then I apologise for complicating the issue.

I hope the info on sound flower is useful to you and maybe other readers, all the same. :)
 
It's still interesting tbh, I don't use a focusrite interface anymore so now I have soundflower installed ;-)

And good luck with the interview!
 
I work as an Freelance Audio Engineer & a huge chunk of my work is as a Podcast Producer, both on my own shows and clients shows.

It seems to me that you are trying to do a lot of things that could be done in the production stage, in the recording stage.

The two most important factors in this are the two voices, yours and the guest. I'm not sure if I took the wrong impression from something you said, but it sounded like you were saying you were not that concerned about the quality of the guests recording. That is the opposite of what you should be aiming for. If you are interviewing someone then for each question you need to think of a ratio between your question and the guests answer of maybe 80-90% their answer. In that case most of the audio you include in the finished show is going to be of poor quality. Try and get the guest recorded as best as you can.

Ask them if they are happy to to record from their end for you. They could use something simple like Audacity if need be. If possible it's always best to get a RAW recording of them in situ and stay clear of using their voice coming via Skype. You then record your voice into the H6 at your end and just use you Mac/mic setup to be the communication device between the two of you. Then all you have to do is think up something you can both do to signify the start, a count in swapping numbers between you followed by something like a clap. Then you just line up the two recordings to that point when you edit. Remember to record both at the same setting so maybe WAV 44.1/16 if not you will end up with the two tracks drifting and you will have to keep "nudging" them to match up again.

As for the clips, edit them back in at the production stage, it also means they will be of better quality.

I hope that helps, make things easier on yourself when recording, then you will have less to worry about at the time, so making you more relaxed when interviewing them. It also means there are less things that can go wrong. You don't want to have to keep saying to the interviewee "sorry, something's wrong here let me fix it". You want them to feel confident in you too, that will also relax them.

Get the two voices right and anything else can be fixed in production :)

Good luck and let us know how it goes.

mcf
 
Back
Top