# Audio Spectrum - how to trim to the selected part?

Hello, I’m connecting Audio Spectrum and I see there are low and high frequencies, how do I take only high spectrum? I see left side and right side are presenting high and low frequencies, how do take the left side only and analyze it?

a trim chop!

You can see that an audio spectrum chop has 22050 samples when you *middle click the node.

Spectrum Chop:
Set the Frequency<->Logarithmic to 0 (frequency mode)
Set the output length to match length to frequency
Now each sample is referencing a specific Hz!

Trim chop:
Unit Values: Absolute
Start and end: the frequency range you want to look at

***Edit oops I put in Right click originally, when it should have said middle click all along!

3 Likes

I’ve tried, but it did not seems working as I expecting,
I want to analyze only selected data and I see Spectrum has left - right = high and low frequencies, I want to trim only from 1-11 in the Trim Chop, but it doesn’t look right in the Trim graph. I assume I don’t understand something or doing something wrong, please help

Ok so the issue is you’ve asked for the hz 1-11, a very small chunk of the practically inaudible low range of human hearing If you only wanted the high end you should use something like 10000, to 20000, which would be way above anything on the high end of a piano.

On a regularity tuned Piano, key 108 has a frequency of 7902.13

en.wikipedia.org/wiki/Piano_key_frequencies

If you were looking for frequencies higher than that, I would go from 7903- 20050

If you were looking for high musical tones I would go lower.

If you looking for strictly half, you could use relative instead of absolute.

You can disregard the numbers you have circled, there are so few samples in that because audio is being sampled at such a high rate. To see what I mean put a resample chop before the trim chop and resample to something like 2000, you’ll see that the numbers you have circled in the trim chop are much higher.

If you want to see how many samples are actually in a chop, you should use a middle click to avoid confusion

Why do I only have 31 as a total number? I don’t understand, I think we got to confusion talking about the numbers. If you look to the Spectrum, the total length of the graph is 31 and I want to take a chunk of it and analyze it. You probably speaking about the Heights of the graph which barely reaches 1. I think I don’t understand something or maybe some basic principles. Would you mind to show me an example of Spectrum base on regular TD samples?

I was using Audio Band EQ - from one of the tutorials to separate LOW, MED, HIGH frequencies, I’ve heard Spectrum does better job, but I don’t get why and how to make it? I want to get High volumes and then analyze them.

OK so you must take into account sample rate when looking at the chop, thats why I recommend the mmb to see how many samples are really there…

In this example, change your master frame rate (FPS on the main playbar) and see how the spectrum chop changes size in the viewer. there are always 20050 samples, just spitting out at different rates based on sample rate
understanding sample rate.2.toe (10 KB)

frames (f) are going to change based on frame rate
samples (i) are not going to

Also check out this post for some uses of the audio spectrum chop to get more complex examples

There are 2 viewers: one is octave based for instruments and singers, the other is a full spectrum viewer.

derivative.ca/Forum/viewtopic.ph … 363#p45934

Get the octave based note compass from that post, Delete the Spectrum viewer… and get the more recent spectrum viewer from the very bottom post in the same thread.

Thank you, I’ve started to understand it slowly.
Okay I have 44100 samples/sec, I want to pick some hight-frequences, where can I get the info about it? What would be the best trim range to cut only the most higher frequencies?

I also going to try TOP Audio, it’s something very new to me

Well it depends on your source!

High is a subjective term!

High as in Musically? For example a quick google search tells me that 27.50Hz – 4186.01 is the range of the average piano, so the very High key on an 88 key Piano would be 4186.01. Do you consider the highest note on a piano high? The recording process will have resonant frequencies and distortions beyond the instruments native range, so you often wont have to look all the way up to the top, just as high as your ear considers high I guess!

I suggest using an audio oscillator connected to your audio device out, and cranking it’s Hz up by ear until you have a value you consider High, and work around that range based on your personal interpretation of high.

en.wikipedia.org/wiki/Audio_frequency

16 to 32 1st The lower human threshold of hearing, and the lowest pedal notes of a pipe organ.
32 to 512 2nd to 5th Rhythm frequencies, where the lower and upper bass notes lie.
512 to 2048 6th to 7th Defines human speech intelligibility, gives a horn-like or tinny quality to sound.
2048 to 8192 8th to 9th Gives presence to speech, where labial and fricative sounds puss.
8192 to 16384 10th Brilliance, the sounds of bells and the ringing of cymbals and sibilance in speech.
16384 to 32768 11th Beyond brilliance, nebulous sounds approaching and just passing the upper human threshold of hearing

Thank you very much, it worked for me!
I also going to try Audio TOP this week!