Some basic sound questions

GoldWave general discussions and community help
Post Reply
Rhondelon
Posts: 2
Joined: Tue Nov 28, 2006 7:15 am

Some basic sound questions

Post by Rhondelon »

I have a few questions about sound editing that may seem simple to some but I've up till now been a strickly, "I'm recording live, problems happen, it's going straight to master and an ounce of prevention is a good start"

Typical systems are a Mackie 24ch live mix to Mackie 8ch digital to hard drive 16-bit signed stereo with less than -90 db ambient line noise direct to CD or a little noisier Mackie 6ch live mix to stereo with around -75 db ambient line noise into a SoundBlaster 128.

Through both of these, I effect some great recordings with the usual artifacts of a live performance, especially when the preacher forgets he's being amplified and the mic ends up listening to the pulpit instead of him

1. Is there a straight forward way to normalize only a single audio pattern (assuming the speaker is like an instrument and has a predictable waveform)

2. I have noticed from time to time, when trying to reduce artifacts (the purpose of getting Goldwave) that there is no way to adjust a compressor/expander/noisegate/EQ over time. I say this because there are sometimes occilations in the environment itself that causes sound qualities to change and normalizing those for me are the most difficult.
Take for example a sound that starts clear becomes a little muffled due to some wind, a sneeze into the mic or other such bothersome non-sense and the live compressor hits hard enough to make any utterations during that event sound like someone talking through tin-can and string (try it, it's a really neat sound)
Okay, now apply this scenario over a predicatable time table (like the fluctuation of a bad motor in a 2" tape drive. Where filter needs to be applied on a bell curve every 1.7 seconds (arbitrary numbers of course). How would one go about this in Goldwave.

3. I've seen several examples of this in here and have the sampled sound filter just about down pat (or more accurately, finding a whole uniform sample of the bad carrier signal) but my question is related to number 2. In recording on my M305 digtal camera in video mode, the condensor in it at high amplification picks up the most god awful digital carrier noise you've ever heard in your life (or neatest sound if your into the matrix scene) The sampled sound filter gets rid of about 90 percent of it which is good enough for webcasting or family but there is this stubborn AM occilation of around 6k that moves from 13k down to 5.5k (approximate) from beginning to end of every recording. It's a great artifact if you wanted to forensically match this camera to a brand / model but very hard to filter out without destroying that whole frequency range through the whole recording (or doing it a second at a time for 3 hours of desk time)

The purpose behind all of this is I am moving to video (insert "You want me to do what? with what?) And need a stabilized audio track that is more suited to playback on high dollar surround systems that aren't going to go blowing our members speakers on a regular basis (or more importantly degrading the sensitive ego's of our audiophiles) All this without moving to an Alesis 24 track recording environment or comparable digital equivalent.

Hopefully I've asked these questions as thoroughly as I can and I appologized ahead of time if the answers are already in the boards.

Thanks,
Brian

"There is no substitute for perfection except the alteration of perceptions" -Some wise guy
DougDbug
Posts: 2172
Joined: Wed Feb 16, 2005 3:33 pm
Location: Silicon Valley

Post by DougDbug »

My first impression is that GoldWave is a bit "light" for your needs. Don't get me wrong... I'm a HUGE FAN of GoldWave. It's a rock-solid program, and the best audio software in the under $100 price range. On top of that, I've never had better tech support from any software company!

On the down side, GoldWave is only 2-channel stereo. I think you need multitrack recording. Also, GoldWave doesn't have a "rubber band" volume control. This makes it difficult to tweek the volume over different parts of a long program. You can use GoldWave's Volume Shaping, but it's awkward unless you are working with a very short segment of the file.

When you buy a multi-channel soundcard, it will come with multi-channel recording software. (Depending on your needs, you may need additional multi-channel audio-processing software.) If your Mackie equipment can't do it and you don't know where to get started, M-Audio sells some affordable hardware.
...especially when the preacher forgets he's being amplified and the mic ends up listening to the pulpit instead of him.
There's not much you can do about that. You can boost the level during that section of the recording, but you will boost the background noise too, and it will still sound "hollow".
1. Is there a straight forward way to normalize only a single audio pattern (assuming the speaker is like an instrument and has a predictable waveform)
If I understand you, no. If you have a speaker/singer with music in the background, you can't process the voice without also processing the music. ...You can't un-bake a cake and you can't un-mix an audio file.
2. I have noticed from time to time, when trying to reduce artifacts (the purpose of getting Goldwave) that there is no way to adjust a compressor/expander/noisegate/EQ over time.
You may have better luck with Auto Gain, which is part of the optional GWVoice Plug-In. You can select a section of the file and apply an effect only to that section. However, there is no way to fade-in or fade-out the effect, so you can get "glitches" when the effect suddenly starts & stops.
Take for example a sound that starts clear becomes a little muffled due to some wind, a sneeze into the mic or other such bothersome non-sense and the live compressor hits hard enough to make any utterations during that event sound like someone talking through tin-can and string
Again once the sound is mixed, there's not much you can do. If the offending noise is comming from a different mic, and it's not mixed with the main track... no problem.
Okay, now apply this scenario over a predicatable time table (like the fluctuation of a bad motor in a 2" tape drive. Where filter needs to be applied on a bell curve every 1.7 seconds (arbitrary numbers of course). How would one go about this in Goldwave.
I don't know of any software that can fix that.
3. I've seen several examples of this in here and have the sampled sound filter just about down pat (or more accurately, finding a whole uniform sample of the bad carrier signal) but my question is related to number 2. In recording on my M305 digtal camera in video mode, the condensor in it at high amplification picks up the most god awful digital carrier noise you've ever heard in your life...
Again, there's not much you can do after the fact. GoldWave's Noise Reduction filter or noisegate may help if the background is at a constant low level. If the camera's AGC is boosting the noise during quiet periods, you might not be able to get rid of it. And, sometimes digital noise reduction does introduce it's own noise artifacts... You need a mic close to the speaker (or sound source) and you need a soundproof studio... :roll:
The purpose behind all of this is I am moving to video (insert "You want me to do what? with what?) And need a stabilized audio track that is more suited to playback on high dollar surround systems...
With live recording, this won't always be possible. There's no software-substitute for good equipment (which you do have), a soundproof studio, skilled "performers", and multiple takes. :twisted:

This sounds like something you are dong on a regular basis... perhaps weekly. This can be very time consuming.You can easily spend several days on post production for one-hour audio program! :shock: At a minimum, you will probably be listening to the program twice, with lots of stops and starts as you make adustments. So, you may have to simply do the best you can in a limited period of time, or just fix-up the obvoiusly bad parts and then call it "good enough".
Rhondelon
Posts: 2
Joined: Tue Nov 28, 2006 7:15 am

Post by Rhondelon »

I appreciate the detailed reply, this has answered many questions about Goldwave's capabilities. I'm going to continue using it but I'll have to look into multi-track. The Mackie digital we have comes with 8 into a proprietary card. I may need to look at more pre-processing equipment as we're only running 2 mildly set dbx compressors on the two main mics. May need to grab a couple 4 channel racks to further reduce 'artifacts'. As far as post-production mix, yea, that's gonna take some time and hands on gettin dirty in the sound room but over time, I should become more efficient at it.

Once again, I thank you for your reply.
Brian

"There is no substitute for perfection except the alteration of perceptions" - Some wise guy
Defenestration
Posts: 24
Joined: Sat Dec 02, 2006 11:43 am

Post by Defenestration »

Rhondelon - You should check out some software called Reaper which is great for multi-track stuff.

BTW, I would have PM'd this to you but the PM facility has been disabled for some reason.
DougDbug
Posts: 2172
Joined: Wed Feb 16, 2005 3:33 pm
Location: Silicon Valley

Post by DougDbug »

I may need to look at more pre-processing equipment as we're only running 2 mildly set dbx compressors on the two main mics.
I assume that the Mackie equipment can record 24-bit. If you do 'go multitrack', that should give you enough dynamic range/headroom to allow post-production software compression on the individual tracks as needed.

Keep in mind that general-overall compression will boost the backgound noise. You wouldn't want to do that with all of the mics at all times. If you are using the compressor as a limiter (probably a good thing) then you are not boosting the low levels & the noise.

Most live recordings need some compression do to playback system/environment limitations. If you leave the live-performance dynamic range, most listeners will end-up constantly adjusting the volume during playback.
OKCBoy
Posts: 110
Joined: Thu Jun 30, 2005 6:42 pm
Location: Oklahoma, USA

Post by OKCBoy »

Defenestration wrote:You should check out some software called Reaper which is great for multi-track stuff.
Wow, Reaper is an awesome bit of software. Thanks for the post.
Post Reply