A newby's questions re: bit rate, sample rate, project rate

[JeffB]

Hi,

I'm no musician or computer whiz though I guess I do have a fairly minimal understanding of terms like hz or bit rate.

I'm looking for a little guidance with respect to recommended settings for recording with Audacity.

I've started recording streaming audio of some college sports broadcasts, and interviews with coaches etc. which is mostly voice, of course, though there might be a little bit of rah rah fight songs, crowd noise etc. thrown in for good measure.

I've also recorded some talk radio and similar teaching materials that are basically just people talking.

I've just been using Audacity's default settings, but wonder if I am wasting disk space, bandwidth &/or computer resources and should use lower bit rates or project/sample rates or whatever when recording those things as opposed to music.

[alatham]

If all you're doing is recording something -> amplifying it (and other light editing) -> and exporting, then go ahead and set Audacity's default sample format to 16.

On the other hand, if you're not saving anything as an Audacity project, then it doesn't matter what Audacity's default sample format is set to. Since you're just exporting it and then closing the program, Audacity doesn't save anything in a 32-bit format.

However, I would leave the sample rate at 44100 if you will put the audio on CDs, and 48000 is you're going to make DVDs.

Moving from 44100 -> 22050 is a much, much larger difference in quality than moving from 32-bit -> 16-bit (despite the fact that the savings in disc space is the same in both cases).

[JeffB]

Mucho thanks, alatham!

My editing is pretty much limited to cutting out commercials &/or deleting things at the end of the show I was recording.

I might do so over several different sessions on different days. When done, I've been exporting to mp3 format. I'm saving them for archival purposes & occasionally to share with people who may have missed the game. Some live out of state and can't pick up the broadcasts or missed them for other reasons.

I haven't been deleting the audacity files but I really need to to save disk space.

So it sounds like I should Edit > preferences > Default Sample Format: 16-bit

............

What exactly is the 44100 sample rate? Does that mean it is recording up to 22050 Hz? I got the impression from another post/thread that halving the sample rate might give the range that was being recorded.

How does the bit-rate fit in? Is that just packing in the same data from the recording more densely?

[alatham]

What exactly is the 44100 sample rate? Does that mean it is recording up to 22050 Hz? I got the impression from another post/thread that halving the sample rate might give the range that was being recorded.

Yes, that's correct. It's called the Nyquist Frequency:
http://en.wikipedia.org/wiki/Nyquist_frequency

How does the bit-rate fit in? Is that just packing in the same data from the recording more densely?

There's a lot of confusion in this little area of the audio world (there's confusion everywhere, but we'll stick to sampling for the moment).

Bit-rate - this number is the amount of storage space it takes to store the audio data divided by an amount of time. Most commonly, it's rated in Kilo-Bytes Per Second (aka kbps). A common mp3 might be 128 kbps, whereas a 44.1KHz 16-bit stereo wav is ~1411kbps.

In order to define bit-depth, you have to first understand what PCM audio is.
http://en.wikipedia.org/wiki/PCM

If you zoom way in on an audio signal in Audacity, it will eventually show you each and every sample. These samples are what your sound card is converting into sound with it's Digital - Analog converter circuit.

So, Bit-depth is the number of bits used to store each sample. It only has any meaning when talking about PCM signals like WAV, AU, and AIFF. When talking about mp3s, bit-depth has no meaning since mp3s don't store audio as samples:
http://en.wikipedia.org/wiki/Mp3

When you talk about re-sampling a piece of audio, you're going to lose all the frequencies that the new format cannot hold. There is no turning back from this, that data will always be missing unless you go back to the original file. I highly recommend not re-sampling.

If you want to save space, and the mp3s you're making are good enough, then go ahead and delete the Audacity .aup files and the _data folders. You'll save lots of space doing that. But once you've deleted them, there's nothing you can do to the mp3 files to make them sound as good as the PCM data that Audacity stores. So make sure the mp3s are what you need.