Audacity Forum

Hi, EarthquakeSam here.

I am working in New Zealand on a project where I would like to be able to play back recorded earthquake sounds which are very low in frequency. They can typically go down to 0.1 Hz and beyond.

Does anyone know if Audacity can play back sound files at many times the recorded speed so that the frequency of these low frequency signals in raised so that a person can 'hear' them in the normal audio range? Kind of like an old reel to reel tape recorder where you record on a slow speed and play it back much faster.

Maybe there is another utility which allows you to do this. Any ideas?

Hi,

There has been a similar question last month here: http://forum.audacityteam.org/viewtopic ... 12&t=10663 which I tried to answer but haven't heard back .

I've never seen actual earthquake data but the other poster said that it is usually recorded at 100 samples / seconds, which would be much too few samples for audio. So, if you manage to open one such data file (using file -> import Raw Data) in Audacity, you could assign 44100 sample rate to it (which is standard for audio cds). Then the content would be automatically speeded up to a listenable pitch. A 1 Hz earthquake frequency would translate to 441 Hz (about the tone A).

Good luck!

Edit: Another thing to consider is the range of value of the samples. If the data samples are floating point they must be in the range of -1 to 1 for them to work as audio. Probably they aren't in that range, so there could be problems. Only the latest Audacity beta (1.3.8) can import such data without clipping. So you probably need the latest Audacity.

After the import, effect->normalize can be used to bring the level down into the range [-1;1].

Edit 2: I got hold of some seismic data. Here's an example with an interesting squishy sound: This is from 3 files in SAC format ending in rwe rwn and rwz (one for each dimension presumably). I put one left, one right and one in the middle to get a stereo effect. imported raw at 44100Hz, 32 bit float, normalized and high passed to get rid of frequencies under 50 hz.

EarthquakeSam wrote:Does anyone know if Audacity can play back sound files at many times the recorded speed so that the frequency of these low frequency signals in raised so that a person can 'hear' them in the normal audio range?

Yes - very easy. use the "Change Speed" effect in Audacity 1.3
If you can't speed it up enough in one go, repeat the effect.

This does the same as the suggestion from george13, but is a bit easier to do.

I'd be interested in hearing a bit if you are able to upload a sample on the internet somewhere.

Yes, right, I forgot about "change speed", good idea!

But the data has to be imported in some way, it's not a sound file. I don't think there is any other way than to use "import raw data" for that. The original sample rate is (from what i have gathered) between 1 and 100 samples per second. If you assign another, higher sample rate instead, like 44100 or 100000 or 200000, you don't need "change speed" anymore.

george13 wrote:But the data has to be imported in some way, it's not a sound file.

EarthquakeSam implied that he already had it as a sound file, but in either case there should be enough information here now.

Quakes are fun. They fill the hole between the bottom of human hearing and battery voltage.

The 1994 Northridge event raised the top of Mt. Wilson (TV Towers for LA) by some value I can't find right this second...and left them there. The noise where I live (south of Northridge and west of Wilson) consisted mostly of books falling off shelves and furniture falling over.

On the other hand, I was standing in the back yard when a 3.9 occurred straight down. Far from being silent, that one sounded exactly like military artillery being fired at close range. The neighborhood cat came around the corner three times normal size.

"Row-wow?"

As has been pointed out many times, Audacity will cheerfully amplify battery voltage if you can figure out how to get it into the computer. Audio amplitude and Richter values can be thought of as cousins of each other for the same reasons. We don't hear and the earth doesn't move linearly.

However frequency translation doesn't always work so good. If a shaker has energy between .2Hz and 20 Hz and you bump 0.2 Hz up to 200Hz, 20 Hz suddenly appears as 20 KHz. You need to be managing a subset of the energy and be mindful of what you're listening to.

What does Audacity do during Input Raw? If it's not "sound" what format was it? My boss is a Cal Tech EE and he can get information and data from friends.

Koz