PRE will automatically down-mix 5.1 to stereo as it does not output 5.1 (only Dolby or PCM). So PRE itself is a good tool to strip the sound from your clips. Go to Share> Personal Computer> MPEG. In Advanced ... uncheck Export Video and ensure Export Audio is checked. On the Audio tab you can then choose between Dolby and PCM. If you select Dolby you can vary the bitrate.
If you are an iTunes user you could also load the clip into iTunes (if the format is supported) and export the audio from there.
Insanity is hereditary, you get it from your children
Another tool, along with Neale's suggestion for using iTunes is the great, free audio-editor, AUDACITY. It can rip multi-channel material, and can Save_As many different formats/CODEC's. If doing this, I would choose PCM/WAV 48KHz 16-bit.
Still, as Neale says, PrE will do the mix-down to 2-channel (stereo) for you.
Now, back in your Project's Preset, you will want to have chosen the 5.1 SS version of the Preset, to simplify the editing process.
As a follow up to the replies to this query - if I wanted to burn a disk with 5:1 surround sound, is there any way of doing it using Premiere Elements, or would I need Premiere Pro or some other program?
Unfortunately not. Even with PrPro, one must purchase the Minnetonka SurCode DD 5.1 SS encoder plug-in for DD AC3 (6-channel) output.
For a DVD, the only 6-channel primary source Audio is DD AC3. Now, DTS and others can be used, but only as supplemental Audio streams.
There was a thread, about a year ago, where a user did a mini-tutorial on doing DD 5.1 SS without the SurCode encoder plug-in. Some seemed to have good luck with it, but some never got that workflow to function properly.
Even if you did, say Export as a WMA (6-channel) and brought that into Audacity, or similar, then encode to DD AC3 5.1 SS, the trick would be to author that to DVD. PrE's limited authoring capabilities cannot do that. One would have to use another program for the authoring. Adobe Encore can author with DD AC3 5.1 SS, but can only monitor it as stereo. All 5.1 monitoring must be done back in PrPro, or the audio-editing app. Still, Encore CAN Burn that 6-channel to the primary Audio Track on the DVD. I think that Roxio can author to DD 5.1 SS primary DD AC3, but am not sure. I do not know about Sony's DVD Architect. Maybe Steve Grisetti can comment, after all, he wrote the book on that program.
Good luck, and sorry for the bad news,
I was surprised to learn that PRE converts the 5:1 sound to stereo, presumably in the final stages since there are five audio channels appearing on the timeline. From this I assume that the most practical way to hear the 5:1 sound (and I don't know how well my camera records it) is to play back the original camera footage via an HDMI cable to a 5:1 amplifier and speakers. I'm quite happy with stereo sound on my movies so I won't be going down the path of purchasing extra software. Many thanks for providing this information.
You are correct in the playback of the 5.1 SS. PrE does do a stereo down-mix for Export.
The reason for this is that Dolby Labs charges about US$ 200 for the license for their DD 5.1 SS. Minnetonka charges ~ US$ 250 for the SurCode encoder plug-in for PrPro, if bought through the link inside of PrPro. Some NLE's offer a version of 5.1 SS, but it is not certified by Dolby Labs, as the Minnetonka SurCode is. Now, there are a couple of NLE's, that do have Dolby certification, such as Sony, but I am not sure what the deal with Dolby is.
As for the 5.1 SS from a camera, this is often not as impressive, upon listening, as it sounds in the marketing literature. Many people end up spending more time trying to "clean up" the signals, than actually editing the video.
Recording multi-channel Audio is an art form and is best done by trained technicians with special equipment, who then do a mix to achieve the depth and placement of aural aspects, that 5.1 SS can. In most cases, 5 - 6 mono mics are utilized, their placement being heavily considered, then their levels being balanced on a mixing board. Five to Six discrete tracks are produced. Often, the LFE track is created in post, rather than being recorded on-location.
With a camera, capturing 5 - 6 channels, the mics are not specialized, are all placed on the camera and just pointed in different directions. Some balancing is probably considered in the design of the camera, but not that much, and it will be common to all locations, and not specialized. Often, the Center channel is just a mix-down from the L & R mics, and a separate mic is not used. What often happens is that the rear-channels pick up unwanted signals from behind the camera operator. These sources are off-camera, and can be horribly disconcerting. Back in film school, the professor doing the sound course opened the first lecture with "if you hear a dog, the audience must see the dog." Obviously, if one is picking up the sounds of children playing in a playground behind a scene with a couple talking in the park, at some times, the "audience must see the children." Now, that "must see" concept was over-stated, but with a purpose.* Much time is usually spent trying to clean up the 5 channels, eliminating much of the rear channels, and often doing more separation on the center channels.
Audiences, even Aunt Marge and Uncle Fred, have become pretty sophisticated, due to the audio in motion pictures. They have been listening to material like THX for many years. Also, audio is much more important, than most casual videographers anticipate. The days, of any audio is better than no audio, are over.
I work in DD 5.1 SS for most of my Projects, using PrPro and the SurCode encoder (often the DTS encoder for supplemental Audio streams), and the multi-track capabilities of Adobe Audition. When I shoot, I attempt to get a good, clean stereo signal, and then a shotgun, or lav. mono mic for any dialog. I can expand the stereo field and punch up the mono Center. This gives me the front 3 channels. I'll then place SFX around the SS "stage," so that they have aural location. Most music comes in as stereo, and I will duplicate that to create the Rear channels, which will be attenuated, and probably have a touch of Reverb and Delay added. The LFE signal is usually created from the necessary signal sources, and heavily filtered, with specific EQ work done. For general work, the LFE channel is usually not THAT important, but I will often place certain SFX into that channel, but it must be clean, or the LFE (subwoofer) will be boosting other frequencies too - something that we do not want. On a few occasions, I have worked with 5 - 6 discrete signals, provided by a sound team, and those are really fun, if the recording was done properly. This is rare for me though.
If I had a camera, that allowed SS, I would do about as I do now - stereo Front with a shotgun, or lav. for Center, and then create the Rear and LFE, so I'd have the SS capabilities set to OFF.
* For me, an example of not "seeing" would be a situation like this: Still lake in the woods at dawn. As the Sun comes up, bird calls echo around the lake. The sound of the wind in unseen trees is heard around the camera. We do not need to "see" the birds, or the trees swaying slightly. Those are aural experiences, that most can identify with. We can hear the birds as we stand beside that lake, but might never actually see them. Unless the wind picks up to get above the ambient noise floor, we seldom will turn to see which boughs are moving. Still, I think that you get the idea.
With the example of the children in the playground, one could perhaps do a wide, establishing shot with the children in the foreground, and the couple in the background, and then cut to a CU of the couple for their dialog. The audience will have seen the children, so having their voices in the Rear channels is now accepted. The audience will know that the camera has now moved in front of the playground to where the couple is standing. One could even zoom past the playground, racking focus from them to the couple, and then cut to a CU of the couple.