Audacity Forum

Hi, as titled, I'm new to Audacity and I'm currently doing my third year dissertation..
It involves Binaural beat stimuli and I am simply looking for some software that will tell me the frequency of the beats and possibly the binaural beat itself (although this can be worked out easily enough
I understand (kinda) that I have to split stereo track to get the right and left ear separate? < Not sure if that's entirely correct but it seems to be..
But the wave is showing as exactly the same on the right and left.. Surely they should be different because the tones are different..
Basically I'd really appreciate anyone who can help me discover what the frequency of my mp3 stimuli is. Would be an amazing help to my dissertation!
Thanks
Jess

As I understand this: You have different pulsing tones on left and right and you need to analyze them? The two tracks will appear the same until you magnify them enough to see the actual waves inside the pulse which will be different.

Finding the frequency may be interesting because you will not be able to use Analyze, Plot Spectrum on very short tones. The best you can do is magnify the waves enough to be able to tell the time between peaks and then use the inverse duration calculation to get the frequency.

f = 1/d

400 Hz wave duration is .0025 sec. or 2.5mSec, and you can derive that from very short pulses of tones.

But the wave is showing as exactly the same on the right and left.. Surely they should be different because the tones are different.

If you zoom-in (in time), you should see a difference. But, I'm not sure if that will help you or not...

It's very hard to analyze sounds by looking at the (time domain) waveform. It can he hard to tell the difference between music and spoken word, and you can't look at the waveform of a song and identify it there is singing or not. (Spoken word is generally less "dense" and there are gaps between sentences.)

I think there's just far too much data... The range of human hearing can go up to ~20kHz, and with CD audio (just for example) there are 44,100 samples (data points) per second. In a "normal' view, where you are looking at several seconds of audio, there are not enough pixels to show all of the data-points. You can zoom-in to see the individual samples, but then you are only looking at a tiny-fraction of a second of the sound. GoldWave has a tool for saving all of the sample-values in a text file, but again there is just too much data to know what you are "looking at".

I don't understand what you are doing, and I don't know what a "binaural beat" is, but the solution might involve creating a Spectrogram, which shows the frequency content over time. Or if you are artificially generating these sounds/beats, you should already know everything about the waveform, and there is little need to analyze or visualize it...

Hi, as titled, I'm new to Audacity and I'm currently doing my third year dissertation...

...discover what the frequency of my mp3 stimuli ...

If your degree is in science, you might look into FFT (Fast Fourier Transform), which is a conversion from the time-domain to the frequency domain, and Matlab (or one of the free Matlab clones). I believe Matlab pretty-much is the standard tool for the scientific analysis of sounds. But, if you are not a hard-science major (or math, or engineering major) you might need some help with it.


P.S.

...what the frequency of my mp3 stimuli is.

For anything "seroius", you should probably avoid MP3. MP3 is lossy compression, and the file has to be decompressed to by analyzed or played anyway. (When you open an MP3 file in Audacity, it gets decompressed first.) A "regular" WAV file essentially just holds a series of numbers that directly represent the sample-values. (Again, it doesn't do you much good to "look at" the bytes in the file, but it's more-straightforward if you are using an analysis tool.)

Binaural Beats: http://en.wikipedia.org/wiki/Binaural_beats

The first example in that article: 300 Hz was played in one ear and 310 in the other.

So you have a stereo track with a 300 Hz tone in one channel and 310 Hz in the other. Make the amplitude of each tone "0.5".
To create this in Audacity, Use the generate menu to create a ton of 300 Hz, then add a new audio track and generate a tone at 310 Hz. Then click on the name of the upper track and from the drop down menu select "Make Stereo Track". You now have a 10 Hz binaural beat. If you listen on headphones you should be able to hear a subtle pulsating at 10 Hz.

Now to do the reverse:
With the above binaural beat, click on the track name and select "Split Stereo to Mono".
Then press Ctrl+A to select all,
Then from the Tracks menu select "Mix and Render".
This will mix the tracks into a single mono track. The beats become very obvious and you can count them:

DVDdoug wrote:you might look into FFT (Fast Fourier Transform)

FFT is not really a suitable tool for analysing very low frequencies as you would need to use a huge FFT size. The only thing that would show up would be the carrier frequency (in the above example, the 300 and 310 Hz tones).