Science Buddies: "Ask an Expert"

Dear Expert,
I am trying to find the correct software to use on my science fair project. I would like to find out if different voices can be identified using digital voice analysis. Please let me know how to proceed with this. I am starting my research today.

Thanks,
Jacob H.

Hi jacob,

This appears to be the experiment you are asking about:

https://www.sciencebuddies.org/science- ... ?from=Blog

This project idea is for advanced science students who can devise their own experimental protocols. If this is not you note they advise picking another project that has more detailed instructions.

Here is a Science Buddies writeup that tells you how to access sound analysis software:

https://www.sciencebuddies.org/science- ... p040.shtml

You may also want to explore the Audacity.sound forge.net software talked about in this writeup:

https://www.sciencebuddies.org/science- ... p029.shtml

Here is a related idea that explores how different people choose their words when writing:

https://www.sciencebuddies.org/science- ... p022.shtml

I hope this helps get you started. Good luck on the project - it sound quite interesting!

Best regards,

Barrett L. Tomlinson

Dear Expert,
The software works. I am stuck on how to read a frequency analysis. I know that it is measured in hertz. Can you help? Do you think I should abandon this project?

Sincerely,
Jacob H.

Hi Jacob,

Do I think you should drop a project where you have already gotten good experimental results? My opinion is NO, but only you can make that choice. I think you should do some research into the background of the experiment. This is really interesting stuff, and I think if you get into it you will be fascinated. I recommend looking at this project writeup:

https://www.sciencebuddies.org/science- ... p012.shtml

Read through the first part of it down to the section called Bibliography, then explore these webpages:

http://www.physicsclassroom.com/Class/waves/

http://www.physicsclassroom.com/Class/sound/

http://faculty.washington.edu/chudler/bigear.html

After you study these pages I would hope you understand that sound is caused by varying air pressure with time. Scientists describe sound in terms of amplitude (how much pressure changes) and frequency(how many times the pressure goes through a cyclic pressure change per second). There are two ways to describe a sound. One way is to graph the pressure as a function of time. A second equivalent but more useful way is to describe frequencies and frequency amplitudes that make up the sound. This is based on the really astounding fact that you can describe any sound pressure versus time graph as a sum of sine & cosine waves of different frequencies. Look at these web pages for some explanations:

http://www.relisoft.com/science/physics/sound.html

http://www.falstad.com/fourier/

http://en.wikipedia.org/wiki/Fourier_analysis

http://www.jhu.edu/~signals/listen/music1.html

The frequency analyzer software does a Fourier analysis of the sound you record. When you speak the frequencies of the sound you emit depends on the exact shape and properties of your mouth and vocal tract. These differ from person to person, so you see lots of changes between people, their moods & stress levels, and the words they say. That is the basis of the experiment you are trying to do.

The background theory on this project is a little advanced for your grade level, so I am not surprised that you are a little confused. I suggest you t ry to understand as much as you can, even if the math is a mystery to you right now. Anything you can understand now will be a huge help later when you study physics later because these concepts are used in the study of sound, light, radio, tv, computers, and lots of other things.

I hope this helps. Hang in there, and let us know if you need more help.

Best regards,

Barrett L Tomlinson