Larger dynamic range without distortion

[ttyR2]

I've been greatly enjoying Audacity. Currently using it to record church music. I dialed back the audio level in the mixer in Vista so that the sound isn't distorting in Audacity but now the level is pretty low in the program. I can do a select-all and choose amplify, and in this one case it applied a '6.0' gain to the whole recording and it sounded great. What control do I need to use in Audacity itself to increase the range more? Right now, it maxes out at about .6 or so.

Thanks for such a great program!

[steve]

but now the level is pretty low in the program. I can do a select-all and choose amplify, and in this one case it applied a '6.0' gain to the whole recording and it sounded great.

That sounds like you have the input gain about right. If your recording peak level is at -6dB, then that should be close enough for a good recording.

What control do I need to use in Audacity itself to increase the range more? Right now, it maxes out at about .6 or so.

I don't understand what you are asking. What "maxes out" and what is it that you want to increase?

[ttyR2]

On the origional recording that had a lot of clipping, the wave form in Audacity wasn't very large but was very clipped. When I back off the Vista sound mixer input level, the clipping goes away, but the sound wave in Audacity stays very small, and isn't using anywhere near the full volume range that Audacity is capable of. I'm still able to add an amplify filter and increase the levels and it sounds great. I'm trying to figure out how to increase the levels inside Audacity during recording so I don't need to use the amplify filter after the fact.

[kozikowski]

Y'all are talking Apples and Oranges. The peak sound meter and the correction tools are in dB, but the blue waveforms are in percent. You can convert the waveforms to match the other tools by BlackDownArrowOnTheLeft > Waveform (dB).

Your object in the capture stage is to get all the sound converted to digital. You need to do whatever is necessary to keep the sound meter from hitting "zero" or all the way up. That will produce unrecoverable distortion.

Delivery of the show is done in post production. The simplest tool for making the existing show as loud as possible without distortion is Effect > Normalize. Let the tool remove the DC levels and my favorite normalize value is -3dB. In Audacity 1.2, you have no choice. In 1.3 you can set the value.

Most people find that straight level setting doesn't do it because live capture dynamic ranges (difference between quiet passages and loud ones) do not fit in the average listening room--say your car.

So then you have to intentionally create distortion with the compression tools. Effect > Compression. These tools have the effect of someone turning down the loud passages and turning up the quiet ones.

The personality of the person doing the turning determines the quality of the show. FM Radio compression is very different from AM Radio compression. Many books have been written on how to do this.

Before you do any of these things to your live capture, I would Export As WAV. This will produce a single, high quality sound file for backup (and copy it to a thumb drive or other location). Audacity doesn't Save sound files and it's not graceful with mistakes.

Koz

[ttyR2]

The origional (undistorted after Vista sound mixer adjustment) recording looks like this:



After trying the Normalize filter as you suggested:



I don't lose any quality by *not* having the initial recording peaking around '0.7' like it does after using the Normalize feature? What I was origionally asking is why can't I get it to look like the normalized image but during the initial recording.

[kozikowski]

Where is the sound actually coming from?

Let me predict the past (pressing fingers to forehead).

You got a feed from the church mixer or you provided your own microphone mixer and plugged it into the side of your PC laptop.

The typical PC laptop has connections for Mic-In on the side. That connection is expecting a sound level roughly a thousand times quieter than the one coming out of a sound mixer or mixing desk. I'm amazed it works at all. Usually, you get clipping and grand distortion no matter what you do in the control panels.

Some laptops have the ability to switch that connection between Line and Mic levels to help with this. Some don't.

Clipping such as you describe generally happens before or at the analog to digital converter. That's the connection at the side of your laptop.

You could switch to a Mac. Mac laptops have High Level Line-In connections and a complete inability to handle microphones.

I've been known to construct an audio attenuator to reduce the level 1000 to 1 before the computer gets it.

One other note. A deskside PC may be a solution. Their sound cards tend to be able to manage anything.

Koz

[ttyR2]

I'm currently using a Dell Latitude D630 and it does query you (at least in Vista) as to what device is plugged into the audio-in jack when you insert a plug. Options include mic and line-in. With the input level slider in the sound mixer (in Vista) set to 10 on a scale of 0 - 100, the sound is undistorted and doesn't look clipped in Audacity.

The recording today turned out well using the Normalize filter, considering it's only in Mono at this point (using a nice Allen & Heath GL2400 mixing console). I hope to do some stereo work during practice one of these days.

I'm not sure if I'm just not explaining my question right, but essentially what I'm looking to do is boost the volume level on the fly *in Audacity* during the origional recording, not after during editing. I don't see a control in Audacity that appears to do that on-the-fly. Right now, the level being fed into Audacity is as hot as it can get without clipping on the hardware.

[steve]

On a mixing desk there are several places to adjust the volume:

1) Input gain - this is the initial signal boost that is applied to the incoming signal. To monitor this you need to select Pre-Fader Level (PFL) for that channel. The level should be adjusted so that it comes close to 0dB but without distortion. Care must be taken if you are running this close to 0dB and then applying Eq. A boost in the Eq section could push the PFL above 0dB, so check the PFL after adjusting the Eq.

2) Channel level - this is the post Eq/Post Fader signal level for that channel and can often be monitored by switching the channel to "Solo".

3) Group - some mixing desks can route multiple channels into groups before mixing into the final stereo mix.

4) Master Fader - this is your main mix level.

You may also have Auxiliary send/returns, but if you are using them then I assume that you know how they work.

Some mixing desks also have a control for the tape out level.

If you are running all of the levels at a "good healthy" signal level, you should have no problem at all in supplying your sound card with a loud clear signal. My suspicion is that you are avoiding having the main PA system too loud by keeping the level low on the master fader, which in turn is supplying a weedy signal to your computer. If that is the case, then the solution is to turn down your power amplifier and turn up the master fader so that you have a strong signal on the main mix.