I recently recorded a video using a remote (wireless) microphone.  I used the option where it records the center "voice" channel and the camera's on-board microphone still records the remaining audio.  (Another option would be use the remote mic as the only audio source, and then it's all in mono.)

The camera records in MPEG format.

THE QUESTION IS, does anyone know a good utility to strip out only one channel (the center channel in this case) and export it to it's own file?

I suspect (hope) the center channel has the best quality of the sound I really wanted to record.



BACKGROUND:

I was recording a graduation speech, and I had my wireless mic on the podium.  I didn't dare use that as the only sound source for two reasons; a) I didn't have headphones along for monitoring purposes, and b) I was afraid it might pick up too much paper rustling or other unwanted podium noises.  On the other hand, the ceremony was in large hotel atrium, and as I feared, the acoustics were poor - the setup had a lot of echo and the hotel left two water fountains running, so there was a lot of background ambient noise.

I figured that using the wireless mic as the center channel would increase my chances of getting good audio of the graduates doing their speaches... especially if I can pick-and-choose between the wireless mic on the podium and the on-camera audio.

The audio is "audible" and at least I can hear and understand the speakers, but I would love to improve it.


Currently my Studio 12 installation is broken, but I know in S11 I cannot mute one channel at a time (as far as I know).  I don't remember if S12 gives me better control.

Does Audacity have any cool "surround sound" abilities so I can mute all but the center channel?

Does anyone have other tools they know of that work (and are low-cost)?

BeSweet is what dun4cheap used to use to do this.

Thanks, BittMann.  I'll check it out.

 

If anyone has additional suggestions, please share.

 

BTW - Here is a link to my wireless microphone.  I've posted here before (in nausiating detail, I'm sure) about how I used it when I recorded theater stage performances at the high school.  I'm quite happy with it.  Read a few of the reviews if you are interested.

Got $300 and change?  Got a few dozen hours to finesse that 15 minute graduation speech?  Then here's one possible solution. 

Here's a dumb, but possibly useful, alternative:

1. Use the on-board mic to capture audience ambience and the amplified speaker's voice.
2. Record some audience sounds, laughter, boos, cheers, or applause when the speaker is not speaking.
3. Use a remote mic to pick up the speaker at close distance.  Ideally, right on the lapel or on the podium.
4. Use #3 as your principal pick-up of the speaker.  Blend it with portions of #2. to create desired effects.
5. Put in elements of #1 as suited.

You can use combinations of the sound and effects track for your mixing, deploying volume, fade, and balance controls.  One can also use the sound tracks of the over-lay track, the voice-over track.  Just drag the sound clip to one of the two primary sound tracks, then cut and paste it to one of the other tracks.  One can also strip away the native sound track and replace it one of the others, or music or effects.

I fear I'm stuck at step #1 - Extracting the 5.1 audio from my video file.

I played around with two utilities:

• Audacity
• BeSplit (part of the BeSweet family)

Here are my results so far (all failures):

• Audacity...
• If I drop my mpg file from the camcorder directly into Audacity, it doesn't play correctly.  I get a bunch of horrible screeching, probably because it's a video file with audio in it, not a plain audio file.  Even so, Audacityonly shows it as being stereo.  I don't see any center channel separation.
• I can create an audio file from Studio.  That goes into Audacity fine, but again it's only stereo, or so it appears.  I don't see any way to isolate the center channel.
•  
• BeSplit...
• I cannot get "BeSplit" to process the original mpg video file.  It generates a zillion "Sync" errors.  Once again I assume that's because it's a video file, not pure audio.
• BeSplit can process the WAV file that Studio makes, but I only get one output (filename01.wav).  I used the "-demux" option, so I should get at least two (one for left and one for right), or hopefully five files (front-left, front-right, rear-left, rear-right, and center).
  

 

People talk about the AC3 file in some forums, but I'm starting with an MPG video file.

I think this is still where I'm at:

• So my first step is to extract the audio from my video file, with the 5.1 surround sound intact.
• Then I can (hopefully) separate the channels and extract or export the center channel to it's own file.

If I could combine the two operations into one, that'd be cool, but for now I'd be very happy if I could get the the desired result at all.

You can use EAC3to.

 

It's a semi command line/gui interface and takes a bit to learn how to use. I don't know what codecs you have on your machine but EAC accomplishes its task by scrounging the various codecs you already have. You can download the ones you don't have. But it will (among other things) convert your AC3 to 6 independent MONO wav files (instead of a MULTI channel wav file as you have generated) Go to the EAC3to AUDIO tab and use the extension WAVS (not WAV)

http://www.videohelp.com/tools/eac3to

 I suggest you also download TSmuxer (free) since it will also work under EAC3to and the 2 programs paired up are incredible for working with AC3, MTS, MT2S...etc

Great suggestion, Jack!

justaviking:

I recently recorded a video using a remote (wireless) microphone.  I used the option where it records the center "voice" channel and the camera's on-board microphone still records the remaining audio.  (Another option would be use the remote mic as the only audio source, and then it's all in mono.)

 

A while back I purchased the new version of this mic (the Sony ECM-HW2 Bluetooth 2 way center channel Microphone) which has 2 way communication built into it. These things really do a heck of a job at completing the full surround sound effect. The cam operator's voice on the rear, the subject on the front. Things passing by can be heard moving from front to back, left to right and so on. The 2 way communication is nice as well. Both the cam operator and the subject wear (suppled) ear buds and can talk to each other. You can switch so that only the subject is heard so you can use the com line as a cuing device as well, similar to how the control room will talk to the news anchor while on air.

Good little mics and well worth the money. It really completes the full and true surround sound effect on my SR12

Jack, et. al.,

So how do I get my "AC3" sound file?  I have an MPEG-2 file.

I tried using my Sony utility to burn a DVD, but it appears that it makes a simple Stereo DVD, not a surround sound DVD.  I used the BeSweet utility and it reports it as being 2-channel audio.  I put in a commercial DVD to make sure I was using the tool correctly, and it reported that as being 6-channel (5.1?) audio, so I think I'm using the tool properly.

I don't know how to properly demux the audio out of the MPEG-2 file that my camera makes.  I tried MPEG-Streamclip_1.2 and was able to extract the audio, but again it appears to be in stereo.

I turned on my camera again, and the display indicated "5.1 Dolby" so I believe the camera is set properly.

 

Jack, what are you using for playback when you can HEAR the "full and true surround effect" on your SR12?

So how do I get my "AC3" sound file?  I have an MPEG-2 file.

I don't use Pinnacle so I couldn't tell you. (The [unamed] program I use accepts MTS/M2TS with ac3 directly onto the time line). I would assume though that Pinnacle would work if you use TSmuxer to extract the AC3 from the raw M2TS clip and then EACto convert it to 6 monphonic wav files (although I'm not sure Pinnacle will accept 6 mono files as opposed to 3 stereo files). Even with that though, Pinnacle still can not produce a LFE track. (I think with BESWEET you're actually producing a MULTI channel wav file. You need to break the AC3 into independent monophonic wav files for Pinnacle)

 

Jack, what are you using for playback when you can HEAR the "full and true surround effect" on your SR12?

I'm burning to Blu Ray and playing back through the PS3 connected to an Onkyo705 surround receiver. The receiver is set for "DOLBY DIGITAL"

 BTW, you keep mentioning "mpeg2 file", what cam are you using?

justaviking;

Can you send me a short 1 minute test clip, I'd like to play around with this to see if I can have any luck. I'm heading down the same path you are, but before I purchased the camera I'd like to know that all of the pieces really work. From the test shots I did on the store I forget to verify that the audio was in 5.1 mode, so the audio side has not been tested to my liking.

Here's a quick visual sample of the AC3. This is a raw unedited clip from the SR12 with the center channel mic. (You will note al tracks are slightly different indicating each mic is picking up something just a bit different)

[click to view image]

My camer is a Sony DCR-SR200.

The files seem to run around 70MB/minute.  If anyone wants a sample copy, and has a good place for me to upload it to, send me a PM.  Otherwise I'd have to stuff it onto a DVD and mail it to you.

 

UPDATE:

I was curious about one thing, and finally did some additional testing.

When I put my video clip into MPEG Streamclip 1.2, it was saying the audio was "128 AC3 2/0".

I took another clip where I did NOT have the wirelss mic on the camera, and now it displays the audio as being "128 AC3 3/2", and I can play a lot more with the various channels and I can clearly hear a difference between stereo/surround/center audio settings.  Sadly, when I try to export the audio from that tool, I only get stereo/mono options.  (That's what you get for free, I guess.)

 

My wirelss microphone has two settings:

• "On" - where the microphone is the sole audio source, and it records in "stereo mode", but it's actually in mono because the mic itself is not stereo
• "Center" - where the microhone is supposed to record the center channel of surround sound.

I am so certain I put in on "center", because then the camera still records the other channels using the onboard mic... in theory.

I will have to do a couple of tests:

• On-board only (surround sound)
• With wireless mic absolutely, positively in the "center" setting
• With wireless mic absolutely, positively in the "on" setting

I am now seriously doubting that the "center" setting works as advertised.  Maybe I screwed up, but I am very sure I used the "center" setting, partly because I didn't have an earphone to monitor what it was picking up, so I figured the onboard mic would serve as a backup audio source.

I'll let you know what I find out from my tests.

As stated, if you use TSmuxer it will not only demux your AC3 from your mpg clip, it will also show you the audio type (and number of channels) the AC3 contains.

On a SR12 clip TSmuxer shows the audio as "AC3 Bitrate 448Khz, Sample rate 48Khz, channels 6"

You will see a similar breakdown for your cam when importing a clip.

Jack Bellford:

justaviking:

I recently recorded a video using a remote (wireless) microphone.  I used the option where it records the center "voice" channel and the camera's on-board microphone still records the remaining audio.  (Another option would be use the remote mic as the only audio source, and then it's all in mono.)

 A while back I purchased the new version of this mic (the Sony ECM-HW2 Bluetooth 2 way center channel Microphone) which has 2 way communication built into it.

Good little mics and well worth the money. It really completes the full and true surround sound effect on my SR12

I second that! Brilliant device and the sound is fantastic.