Hi,
I Cant find exactly what I’m looking for in FAQ and other posts so I had to start a new topic. Total newbie so please be kind.
I have imported a recording from a cassette tape. It has vocal speech with a background audio track. I have exported the file to Mp3 and played it back and am unhappy with it. There is white noise when there is no vocal speech. What tools/ methods do I need to fix it? I have no idea what the problem is so any help would be greatly appreciated.
Thanks
I just tried noise removal at 15dB and it sounded better on the PC but when i play it on an external device it is hissing when there is no verbal speech. It is even hissing when the background audio track is playing.
I have exported the file to Mp3
First problem. Never do production in MP3. The process of making an MP3 creates sound damage and makes filters, effects and tools more difficult to use. Export everything in WAV until you get to the point where you want to upload a show or put the work on your iPod. Then use the MP3.
There is white noise when there is no vocal speech.
Actually, it’s there all the time but it’s more obvious when people stop talking.
Noise Reduction, but don’t bet the farm on it working well.
Noise Reduction works in two steps. You expose Noise Reduction to a sample of the evil sound, say for example, a ventilation fan whine. It needs to be the fan noise absolutely by itself. That becomes The Profile. You launch the tool again and apply it to the show and Noise Reduction will try to subtract that exact ventilation fan noise – the Profile – from the show. So far so good. You can imagine those steps without too much trouble.
http://manual.audacityteam.org/o/man/noise_removal.html
When you make a Profile from a sample of tape hiss, Noise Reduction gets a profile of “All Sound.” I’m not making that up. White Noise has tiny samples of every sound you can hear. When you apply Noise Reduction to the show, Noise Reduction will try to remove – the whole show. Obviously that’s not optimal, so there are tricks you can try. Frequency Smoothing is a setting that causes Noise Reduction to “go around” spoken words and doesn’t try to do anything to them. If you have a simple, well behaved show, sometimes that’s enough. If you have severe hiss, this process will give you hissy words over a quiet background. It turns everybody into snakes. If you have a complex, highly produced show, it doesn’t work at all.
It helps to avoid the idea you’re going to turn this into a studio production. That’s not going to happen. Start with modest noise reduction values like 6 dB or 12 dB and work up. Each time you increase the reduction, listen for the hiss reduction, but more important, make sure nothing bad happens to the voices. You are looking for honky, bubbling, gargling, fading voices like a bad cellphone call. That’s what it sounds like when you apply too much Noise Reduction.
Nobody will be surprised if all you can do is gently reduce the hiss and still leave the voice performance in pretty good shape. Like earlier, it’s never going to be Glen Glenn Sound Studios. Recovering from hiss is very difficult.
Koz
Yes, there is the problem of bad sound system hiss when you’re playing nothing at all. Nobody is throwing awards at Windows soundcards for their high quality. Windows PCs are not Digital Audio Workstations. Establishing your sound system is high on the list of things to do before you try to do post production on a show.
With the possible exception of headphones, you’re not likely to be holding your mix-down sound system in one hand, either. Nobody should be doing sound analysis on built-in laptop speakers. How do you know when you get it right?
Koz
First problem. Never do production in MP3. The process of making an MP3 creates sound damage and makes filters, effects and tools more difficult to use. Export everything in WAV until you get to the point where you want to upload a show or put the work on your iPod. Then use the MP3.
When you say "then use the MP3, do you mean open the WAV file in audacity and then export the WAV file to MP3 from there?
What Koz is saying is to do all of your processing with files that are WAV format or, better still, Audacity’s native floating point format. Don’t try to export as MP3 until the show sounds good or as good as you can make it.
What sort of “background audio” does this recording have? Music? When you say there is “There is white noise when there is no vocal speech” do you mean the noise gets louder, or is is just that you notice it without the voice covering it up?