How do I record PC mic/headset and PC sounds simultaneously?
 

Granted, this is not really a field issue, but Im developing podcasts with finish work in Sony Vegas and Im running into a terrible issue and no one as of yet has been able to solve it for me.

Simple problem but its proving to be a difficult solution. I produce podcasts on my Blackbird and collect voice audio from my mic/headset with its integrated sound card in the USB connection. Works fine and I record with any number of programs.

I also have about 50,000 sound effects, foleys, and bump audio segments on my PC. I can record these fine by themselves if I wish.

My hurdle is recording both my mic and the sound effects on my PC at the same time using one audio program. When I choose the sound recorder device for one, Im locked into it for the other and its no good for my second need.

Adding the sound effects in post production is not really an option because this is recorded "live" so to speak and is very informal and my sound effects I add from my PC will be very unscripted and must land precisely with my microphone audio to pull my podcast off.

OK, being on Vista, is there anything I can do to get this accomplished by itself or does a program I havent used support this? I tried Audacity and could not get it to work. I used to have Acid but have not tried it there.

How do heavy Podcasters accomplish this dilemma? It would be a hurdle most of them would have to overcome I would think.

Any help, direction, suggestions would be greatly appreciated. Its set me back a good week in development. Thanks.

Let me see if I understand what you want to accomplish. It sounds like you'll be recording/playing a live mic feed, and at the appropriate moment you want to hit a button or click a mouse and play some sound effect that returns to the PC and gets recorded/played together with the live mic feed.

Is this what you're wanting to do?

One obvious thought would be to have the sound effects on a separate PC so you could merge them with the live feed, but I'm assuming you want this to all happen on one PC.

What other gear do you have that might possibly be part of the solution? Or do you want to do it just with the single PC?

Yep, I want to do exactly what you are asking.

I do have a laptop as well, but all my digital juice sound effects are on my PC so Id rather do it all from one PC if I can.

One quick question - how do you plan to "cue up" the sound effects/what will you use to play them?

Not sure I can figure it all out either, but I have a couple of ideas about things I'll try to track down tomorrow.

I plan to just click them with my mouse on my PC as Im yacking into my USB headset.

Hey Chris,

If you have any of the sound blaster cards you will be able to do it.

pull up the sb control panel, far right side(record)...change to "what you hear"

On the other inputs, select your sources,(mic, cd audio, etc) then record in your editing suite.

Me, personally, I would record 2 tracks, but do it how you want to.

Hope this helps.

The software most sophisticated about picking up various streams in your PC for recording is Total Recorder. I don't know if it does the trick you're looking for, but you could find out with a trial, an email, or a phone call.

Also note the digital mixer add-on for their Professional Edition.

No I dont have a Soundblaster card. I tried Total Recorder but could not get it to work.

HOw do musicians do this? They oftentimes play tracks on their PC while inputing an instrument. Something is recording both at the same time.

Would a product like Cakewalk do it?

The only scenerio in my head I can wrap around to get it to work thus far is Getting a Y-Cable for microphone input and running my headset mic into one connection and my laptop's output into the other? Are there Y-Cables for microphones and wold the line-out on the laptop work ok into the mic input on the PC?

Id rather not use two computers if I dont have to.

Seth,

I had thought about something like this, but Chris made the comment that his headset connected via USB so it wouldn't flow through the sound card and as far as I can tell, wouldn't be picked up by Total Recorder. I might be wrong.

The other little "gotcha" as I see it is that he wants to fire off a sound effect through whatever player he uses while talking. In other words, not just dropping the SFX file into a timeline.

I did verify that I can launch an audio player and play a file through a sound card to my speakers while simultaneously recording in Vegas/Sound Forge.

The trick is how to send the output of the player out to "something" and merge it with the mic input. I think a Sound Blaster would do it, I know I could take a sound card output into my firewire interface and "mix" it with mic input and record the two tracks, butwithout an external interface of some kind (and with a USB mic) I'm coming up empty.

Chris,

There are a bunch of ways to do what you want and they aren't so complicated, but your particular setup doesn't seem to have the parts that would make it a piece of cake.

If your mic were not a USB mic I think any program that would record everything going through the sound card might do the trick.

In essence what you need to do is take an output signal from your PC and mix it with your mic input and record the mixed track (or tracks) back into the PC. The big question I have is how/where to pull off the mixing with the setup you describe.

I can also hook my headset up manually, it just has a USB adaptor I tend to use. So if I connected both jacks (mic and earphone) to the card, then how would I solve this issue?

A good introduction to how musicans do it is the DVD "Roger McGuinn's Guide To Home Recording on a Computer" that you can find from a number of sources.

The way multitracking is usually done is they usually lay it down in multiple passes. Multitrack programs like Vegas or Audition can playback one or more tracks while simultaneously recording a new track. Using a program like that you lay down your first track start to finish. You set your cursor back to the start, set the newly created track for playback, and arm the software to record a new track. When you hit play, the original track plays back through the output while the new track is being recorded through the input. When that pass is finished, you rewind back to the start again, arm yet another new track for recording while playing back the first two tracks, and so forth. The audio from the previously recorded track(s) isn't usually recorded into the next, although it could be mixed over if you wished using an external mixer - I've got mine setup so that I can patch the playback to mix into the input if I wanted to for some reason. Instead you get multiple parallel tracks - drums, guitar, bass, vocal, etc - that can then be mixed down into the final stereo track in your postproduction sessions. It could work the same way for a voice track, music track, and FX. Podcasts are not usually broadcast in real time. What is wrong with adding the effects, bumpers, etc in post? Lay down your voice track first, insert the FX on a dedicated FX track that is in parallel with the voice track by dropping them into the timeline, do the same with music on its own track, and mix. The reality of production is it takes a lot of planning and repetition with meticulous attention to detail to end up with a piece that sounds completely relaxed and improvised off-the-cuff.

If you want to do it in real time you need a full duplex, multichannel audio interface of some sort. Full-duplex means it can playback through its outputs at the same time it's recording through its inputs. As well, it needs to have multi-client drivers that permit multiple applications to use it simultaneously. Usually each application running would be associated with a own dedicated set of outputs reserved exclusively for that application's use but a single set of inputs could routed to mutliple applications for recording. You can cue your FX etc through one application for playback, send that output into a small mixer when it joins your mic signal, and then back into the computer through a set of inputs routed to the recording application. For example, I just did a fast experiment taking Real Player and putting it on a net radio station playing some music out through my interface. That output pair normally goes to an Aux input on my Mackie desk and then to my speakers through its control room outs but I can also send it into the main mix. Meanwhile I have a mic hooked up to another input channel also routed into the main mix so the signal at the Main Output of the mixer contains both the playback from Real Player and my voice through the mic. The mixer feeds its mains to the inputs on my Echo firewire interface. I fired up Audition, armed a track for recording from the inputs the mixer is feeding, and hit Record. The result is a stereo track being recorded in Audition that has the music from Real Player mixed with my voice coming through the mic. The tools i used - Real Player, Audition, Echo AudioFire 12 interface, Rode NT1a mic, and a Mackie 1642 desk. That setup has far more channels than you need for your project but a similar assortment of tools and you'll be set.

The problem is the sound effects are totally unscripted. I may want to call to them 3-4 times as laugh lines or to make a point about something. Timing and sequencing with my voice audio is absolutely paramount. I cant "plan" this laugh line beforehand or gag or funny sound effect. It just happens or it doesnt happen as the broadcast unfolds. You cant just go back later on and drop these in because in order to do that I have to "cut" and "pause" to make these holes in the audio in order to get these bumpers in. Plus I will have no idea how big or small to make these holes anyway. Its just way too imprecise. Its just not how you do a quality broadcast in my opinion and not how live radio works (which I do). This is live radio thats recorded. Thats how I need to approach the recording anyway.

It sounds like you're thinking single track, not multitrack, recording like you'd do when recording an aircheck of a live show. But since the show won't be listened to in real time you're not stuck with the compromises inherent in live production and you have the opportunity to refine it in post. You don't need to leave a hole to insert the effect later or otherwise have to precisely time the voice, FX, and music on the fly. You can go ahead and record your voice track as a continuous performance. Let's say your (admittedly hokey) line is "On a dark night in old London Town the sound of Big Ben evokes memories of Jack the Ripper prowling the fog years ago. Come with me back to those dangerous days...." This is recorded on one track. On another, parallel, track you insert the FX of chimes in the distance positioned so the first strike of the bell coincides with the start of the words "Big Ben," with the level ducked down under your voice. As your voice finishes the words "...years ago" you fade up the chimes for the balance of its length. You also slice the voice waveform after "ago" and slide the remainder of the track starting with the words "Come with me..." down in the timeline until they're coincide with the ringout of the last bell peal, filling in the gap with roomtone so it's not audible that the track was cut. Voila! perfect timing without having to worry about it, even think about it, as you actually recorded the voice portion.

At any rate, my previous post has been edited to discuss how you can do it the way you're asking. I've just now done it, recording some dialog in Audition will real-time mixing in FX from my library played back in Real Player and in Winamp, so I know it can work if you really want to take that approach. I can't imagine delivering a presentation while simultaneously searching through a 50000 cue library for the right cut so I could play it at the right time, but the recording setup itself works like a champ if that's really the way you want to go. I've a couple of years experience as on-air talent in live radio myself and IMHO to pull it off you'll need a team with someone else handling the playback and engineering as you perform.

Chris,

As usual, Steve has given some good advice and he certainly knows more about all this than I do.

But what comes through to me from your responses is that you want your voice itself to react to the sound effects that you've heard so your voice and the effects play off against each other. I'm guessing that you don't have a mixer or any other such device. I'm also guessing that regardless of how many sound effect clips you have in total, you have a pretty good idea which few you'd really use and you can put their icons on your screen so you can click them when you want.

Am I at least close?

If so then the only issue is how to get two audio streams to merge.

I haven't actually tried this, so it may be totally unsuccessful, but what I was thinking was that maybe you could run both a line in signal (from an ipod or any other player) into your sound card's line in port and at the same time run a mic into the mic port and see if you hear both signals playing concurrently. If so, you're about 70% of the way there. I'd try it myself but I'd have to move a lot of stuff around to try it and am a bit under the weather at the moment so can't do it just yet.

Anyhow, it may be a long shot, but if it works, you could copy your sound effects to your notebook and run its line out port into your PC's line in port, and run your mic into the PC's mic port, and run the sound effects from the notebook and maybe get the effect you're after.

Again, just a wild a-s idea but maybe worth a try.

yeah you understand exactly what im trying to do. i dont have any special breakout boxes or mixer hardware right now. not sure i want to invest in it either for this.

i looked at my laptop and didnt see a line out. just a video out. could i use the microphone port as a line out or will that not do it?

worst case I think what I may do is run the podcast with my headset mic and queue the sound effects on my PC which will still come through my headset earphones. I will click the sound effects as if everything is recording (both items) and so the timing will be exact. Then go back in Vegas and add the effects in at that time. At least doing it this way, the pauses -- if any -- in my voice or my reactions to the sound effects will actually be real and improvised like a live broadcast. I just then need to drop them into their appropriate spots in post production as I used them during the live recording when they were used but unrecorded.

Probably my most painless solution among a host of solutions?