Advice for audio beginner, please

[Mark Fry]

I'm looking for some "best practice" advice for the audio side of my railway video productions. I described my set-up at some length when I joined dv-info last year, so I won't repeat myself here.

For monitoring, I have a TV monitor (old Philips 8853 with two speakers built-in), 4 little speakers and 1 small bass-speaker that came with my Audigy Live sound card, or a pair of Beyer Dynamic headphones, which I guess are my best bet. With very little effort, I could move a hifi amp and a pair of small KEF speakers into my den. The room is pretty dead, acoustically - lots of shelves, lots of clutter, few hard, flat surfaces. What I don't have are proper near-field monitor speakers.

On my last production, I spent a long time manually adjusting the volume level of each clip to keep it below -6db, following some advice I'd read about DVD audio. It took longer than the rest of the editing and DVD-authoring put together. Was this worth doing, do you think? Surely there's a better way?

The main weakness of my NLE (Liquid Edition 5.62) is a lack of audio tools: no compressor (I've been warned off the "maximiser" tool), only rudimentary echo and EQ, and no support for plug-ins. I know this has been addressed in the new version (LE6), but there are too many bugs in that at the moment, and besides, I need some hardware upgrades before it will run well on my PC. If I'm going to do anything other than volume adjustments, I need to export to another program. I have recently down-loaded the demo version of GoldWave for this purpose

The live sound track is basically the sound of the trains approaching, passing and going away. Sometimes the camera is right beside the railway line, so the camera's ALC (the XM1 does not have manual over-ride) is reducing the volume a lot; sometimes the trains are quarter of a mile away, e.g. across a valley, and their sound hardly registers. Normally, they pass about 20-100 yards from the camera. The important thing is that it should sound completely natural.

Here's what I think I probably ought to do:
1) In LE, finish all the editing, boost or cut the volume at specific places where needed (e.g. when a car passes out of shot), and add sound effects (not usually necessary for my stuff).
2) Export the sound to GoldWave as a 16-bit, 48kHz .wav file. I'd probably do this as a few 15- to 20-minute chunks rather than one complete 75-minute file (that's how I usually structure my productions).
3) Add EQ and/or compression to compensate for microphone characteristics and camera ALC. I presume that this is the bit that needs most skill, and where good advice will be very much appreciated. One thread I saw around here recommended some settings for "gentle" compression - threshold = -19db, ratio = 2:1, 0.1 msec wait-time (?). Does anyone have any favourite EQ settings for the MKE300 that I could try as a starting point?
4) Do I then need to "normalise" the audio to a particular level, and if so, what level and how? My final output formats are DVD (MPEG2) and miniDV tape (will also be used as a master for VHS copies).
5) Re-import the processed .wav file to LE and replace original audio tracks.
6) Dump to master tape, make DVD, distribute to grateful public...

I want to put music under the intro and end sequences, reducing the volume of the "live" sound track accordingly. Should I do this at stage 1 or after stage 5? I think it should be at stage 1, but maybe not include the music in the exported .wav file, only the reduced live sound-track. Does that make sense? What about a voice-over track? Should it be processed separately or together with the live sound?

Does anyone have much experience of GoldWave? Is there much to choose between it and Audacity? Is there something better at the free/very cheap end of the market. It was suggested to me that paying for a better quality VST plug-in would be better value for money than paying for an editing program per-se. Is this good advice? If so, what plug-in(s) are recommended for my sort of work?

What I really want to avoid is messing up the sound, which is why, so far, I've not played about with it. However, if I can make it better...

I know I'm asking a lot, and maybe I should buy a book (Jay Rose is mentioned often around here). I've been looking for FAQ pages and the like, but not found much - if you know of something that covers my questions, please point me at the right the sites.

If you've read this far - thanks for your patience. I look forward to reading your replies....

[Pete Wilie]

Mark,

I'm pretty much in your shoes, still on the steep audio learning curve. :)

But I found "Great Sound for Digital Video" by Jay Rose to be very helpful. You can check it out at his web site http://www.dplay.com/. He has a link there where you can get it for a discount at amazon.com for about $35 US, but that my not be of much help since you are in the UK.

This book is one of the best written books for quick learning I have come across. He addresses your urgent questions right away, and then teaches you how to get great audio while only going into theory as little as possible. His writing style is easy to understand.

Good luck with your projects. HTH.

Best Regards,
Pete

[Ty Ford]

<<<-- Originally posted by Mark Fry : I'm looking for some "best practice" advice for the audio side of my railway video productions. I described my set-up at some length when I joined dv-info last year, so I won't repeat myself here.

For monitoring, I have a TV monitor (old Philips 8853 with two speakers built-in), 4 little speakers and 1 small bass-speaker that came with my Audigy Live sound card, or a pair of Beyer Dynamic headphones, which I guess are my best bet. With very little effort, I could move a hifi amp and a pair of small KEF speakers into my den. The room is pretty dead, acoustically - lots of shelves, lots of clutter, few hard, flat surfaces. What I don't have are proper near-field monitor speakers.

On my last production, I spent a long time manually adjusting the volume level of each clip to keep it below -6db, following some advice I'd read about DVD audio. It took longer than the rest of the editing and DVD-authoring put together. Was this worth doing, do you think? Surely there's a better way?

++ recording them that way to begin with, or using the mix automation feature in a nonlinear audio editing system instead of adjusting each clip.++


The main weakness of my NLE (Liquid Edition 5.62) is a lack of audio tools: no compressor (I've been warned off the "maximiser" tool), only rudimentary echo and EQ, and no support for plug-ins. I know this has been addressed in the new version (LE6), but there are too many bugs in that at the moment, and besides, I need some hardware upgrades before it will run well on my PC. If I'm going to do anything other than volume adjustments, I need to export to another program. I have recently down-loaded the demo version of GoldWave for this purpose

+++ Always be very careful when using software you're not familiar with+++

The live sound track is basically the sound of the trains approaching, passing and going away. Sometimes the camera is right beside the railway line, so the camera's ALC (the XM1 does not have manual over-ride) is reducing the volume a lot; sometimes the trains are quarter of a mile away, e.g. across a valley, and their sound hardly registers. Normally, they pass about 20-100 yards from the camera. The important thing is that it should sound completely natural.

+++well, you already lost that with the camera alc. And as you've found, the trains coming across the valley are too far away. Abandon "completely natural" for "convincing."+++

Here's what I think I probably ought to do:
1) In LE, finish all the editing, boost or cut the volume at specific places where needed (e.g. when a car passes out of shot), and add sound effects (not usually necessary for my stuff).

+++OK+++

2) Export the sound to GoldWave as a 16-bit, 48kHz .wav file. I'd probably do this as a few 15- to 20-minute chunks rather than one complete 75-minute file (that's how I usually structure my productions).

+++OK+++

3) Add EQ and/or compression to compensate for microphone characteristics and camera ALC. I presume that this is the bit that needs most skill, and where good advice will be very much appreciated. One thread I saw around here recommended some settings for "gentle" compression - threshold = -19db, ratio = 2:1, 0.1 msec wait-time (?). Does anyone have any favourite EQ settings for the MKE300 that I could try as a starting point?

+++There is absolutely NO way to communicate what setting may be helpful without hearing the sound. Am I remembering that you don't have any audio monitors, but some small KEFs? This is courting disaster. Especially if your intended playback system is anything close to theatrical.+++


4) Do I then need to "normalise" the audio to a particular level, and if so, what level and how? My final output formats are DVD (MPEG2) and miniDV tape (will also be used as a master for VHS copies).

+++Not necessarily+++

5) Re-import the processed .wav file to LE and replace original audio tracks.

+++OK+++

6) Dump to master tape, make DVD, distribute to grateful public...

+++OK, maybe+++

I want to put music under the intro and end sequences, reducing the volume of the "live" sound track accordingly. Should I do this at stage 1 or after stage 5? I think it should be at stage 1, but maybe not include the music in the exported .wav file, only the reduced live sound-track. Does that make sense? What about a voice-over track? Should it be processed separately or together with the live sound?

+++At some point, after cleaning up the tracks, you have to do a mix. That's when you combine everything. Probably during steps 3-4-5.+++

Does anyone have much experience of GoldWave? Is there much to choose between it and Audacity? Is there something better at the free/very cheap end of the market. It was suggested to me that paying for a better quality VST plug-in would be better value for money than paying for an editing program per-se. Is this good advice? If so, what plug-in(s) are recommended for my sort of work?

+++I can't help. I use Pro Tools.+++

What I really want to avoid is messing up the sound, which is why, so far, I've not played about with it. However, if I can make it better...

++You may muck it up. My advice is to find someone to help you. I did this for a friend a few years back. He put the piece together and said he'd call me when it needed a tweak. I saw him (and it) at a local informal screening. The levels were erratic, sound wasn't in the right places, so the audio was distracting. I said, "So you want me to tweak iit now?"

He said, "Please!"

It took about 3 hours to go over the airplane dogfight and trench warefare scenes. The entire piece was about six minutes.

http://www.bizclip.com/herrondesigns/afallfromtheclouds/

As it turns out the short has won a nimber of awards, including several for the soundtrack.+++

If you've read this far - thanks for your patience. I look forward to reading your replies.... -->>>

+++ Best of luck+++

Ty Ford

[Mark Fry]

Thanks Ty & Pete,
I spent some time at the weekend messing about with GoldWave, with generally unsatisfactory results (due to my inexperience, rather than deficiencies in the program, I'm sure), so I'll leave well alone for my current projects. I will save up for some proper monitor speakers, read all I can in the mean time, and keep experimenting on things that don't matter so much.

[Bruce S. Yarock]

My back ground is in songwriting and music production, and I've been using Cubase and Nuendo ( Steinberg) for several years. When I started doing video a few years ago, I realized that the best and easiest place to do the audio editing was in these programs.
When we finish editing the video, we bring the video clip into Nuendo. (n Ther you can view the track while working on the audio). Then we take each of the audio tracks ( in Premiere) and create a wave from the beginning to the end of the piece. Import each wave to a seperate track in Nuendo, do ANY editing I want, and add any additional tracks I want.
Last step is master the audio in eaither Wavelab ( or soundforge), Import the finished stereo wav back into Premeier, and ctreate the dvd.
My girlfriend was trying to edit audio in one of the NLE progarams, and it was ridiculous. It's a piece aof cake in a real music recording program.
Good luck
Bruce Yarock

[Bob Costa]

<<<-- Originally posted by Bruce S. Yarock :
My girlfriend was trying to edit audio in one of the NLE progarams, and it was ridiculous. It's a piece aof cake in a real music recording program.
Good luck
Bruce Yarock -->>>
Too bad you didn't choose Vegas. It used to be an audio editor.

[Mark Fry]

Bruce - what does your mastering step involve? Why do you do it in Wavelab or Soundforge, rather than Cubase or Nuendo where you have done your editing? I'm not clear what the relative strengths and weaknesses of these audio programs are.

Thanks

[Bruce S. Yarock]

John-
i think Vegas is great ( I'm trying to learn basic editing on it), but I've been using Steinberg for 7 years, and find it real easy. Also. I tried to do some slicing and dicing in Acid, and I found it very un intuitive ( compared to Cubase). Since vegas works similar to Acid, I've had a mental block against it....

[Bruce S. Yarock]

Mark,
i've always used Wavelab for mastering because-
1- I have all of my plug ins and presets saved there,
2- I can look at the mixed down wave and trim, dc offset, normalize, and any one of several other edits, which would be impossible to do BEFORE mixdown. It's really not too hard, once you get used to it ( and I'm no tech genius). I usually trim thr wav,do dc offset, normalize and tweak anything else. Then I do some minor eq, etc. the last step I use is " Loudness maximizer", which is a nice mastering compressor, and I'm done.
3-You can do all kinds of sound restoration in Wavelab that would be hard in the multitrack programs.
After all is said and done, I like finding something that works, and staying with it, till I get dragged ,kicking and screaming, to the next upgrade or new "miracle" program.
Bruce yarock

[Jay Massengill]

While ACID and VEGAS have a similar interface, they are very different (I feel) in how they operate during editing.
I'd keep working with VEGAS until you're more comfortable. And especially make certain you know the commands that are available. It is very fast and easy to use with practice.

[George Ellis]

I use LE 6. I do not compress my audio, but I do edits, cleanup, and other audio tweaks outside of LE (more tools in 6, but still not up to snuff).

I mark in and out and Export my audio as WAV. I then use Sound Forge 7 to do any post work I need (usually Noise Reduction), save it, and then reimport it into LE6.

Sony offers a free trial of the software on their site. Videoguys sells SF7 with NR2 and CD Architect for a great price.

[Jeremy Davidson]

I use Goldwave to do a couple of final processing steps on my audio projects (mixed and produced in Cakewalk Sonar). Mostly I just use Goldwave to normalize the final mixed WAV and tack the proper amount of silence to the head and tail (for CD mastering).

I have used it on occasion to do music loops and edits for video. I haven't used its compression tools much. Note that it does support DirectX plugins now.

I also used Sonar for a wedding where I multitracked the audio coming from different mics and aligned the mixed WAV to the video in Premiere. It worked, but it took a lot of time.

I've not been impressed with the audio functions of Premiere (version 6.0). At most I'll do some rubberbanding for volume and maybe a high-pass filter to roll off some mic rumble or such.

Yes, I've heard that Vegas is better -- maybe next time.