-
WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
how is it interpreted?
- Thread starter musica
- Start date
- Joined
- Oct 25, 2019
- Messages
- 11,914
- Likes
- 16,333
Depends what you are looking for or trying to understand. Where is that specto from and why are you looking at it ?
1) The recording/speaker rolls off around 35Hz as the content below that is likely noise.
2) between 40-50 Hz there are peaks. Either the instrument on the recording has this as a fundamental, or this is a peak in the room that is exaggerated.
3) There is a gap and then a restart. There is no noise is this gap, so if this is a mic recording, there are dropped frames or if this is a digital analysis of a recording, we just want from one track to the next
Without some boundary conditions or info, there is very little more that you can comment.
For example:
1+1=2
What can be learned from this? It could be a simple fact that one added to one equals two. Maybe the plus in in blue because this is supposed to be a first grade test and someone has filled in the blank and the blue is what is written by the user. There are different fonts and sizes. Maybe this is just a test of HTML rendering and the forum software, etc.
- Joined
- Oct 25, 2019
- Messages
- 11,914
- Likes
- 16,333
What the sounds/ music looks like in terms of frequency and amplitude over time. So in the one you showed you have around 5 minutes of signal with low volume sounds between 20hz and 95hz but a couple of strong tones at 45 and 50 hz (the yellow) going pretty much throughout. And a 20 second silence around 4min 40/ 280 secs.
But you are only seeing the lowest 100hz of a 20000hz audible band- so not even all the bass. So its only telling you what the sonic content is in the bottom 3 octaves (and looking at the 95hz cutoff, perhaps thats all there was there anyway)
- Joined
- Oct 25, 2019
- Messages
- 11,914
- Likes
- 16,333
We guessed that from the tag on the spectro- but its not the whole song- its only the bass and Im not sure its actually that even.
So where did you get the spectrogram from?
- Joined
- Oct 25, 2019
- Messages
- 11,914
- Likes
- 16,333
If you want a tutorial on doing spectrograms using SOX and interpreting them, then ask that. If you have a particular goal in mind- state it. Your OP is at best vague and open ended and could be construed as time wasting.
And please- do your own web research before coming here to ask questions.
And please- do your own web research before coming here to ask questions.
It depends what the spectrogram is of.
You have to know what is measured, how it is measured, as well as the result of the measurement (spectrogram) if you want to draw correct conclusions.
What you are doing is like showing a random waveform and asking for what can be understood, without knowing if it is a waveform showing (for examples) the speed of a car, or seismic vibrations.
EDIT - now you have shown it is a song, but the spectrogram is incomplete showing only frequencies up to 100Hz, and with a gap at around 270 seconds. It is not very useful as analysis of the songs spectral content.
- Joined
- Oct 25, 2019
- Messages
- 11,914
- Likes
- 16,333
Nor do we have any idea why OP is (trying ) to analyse spectral content despite at least 3 separate requests .
- Joined
- Oct 25, 2019
- Messages
- 11,914
- Likes
- 16,333
Then you need to figure out how to take (or show ) a full range 20 hz to 20khz spectrogram. I don't even believe the one you posted is a correct representation of vogue below 100hz (there isn't 2 constant high level 45 and 50 hz tones in the song )
solderdude
Grand Contributor
Last edited:
A spectrogram shows where the energy is distributed in the frequency range (the 'spectrum').
Brighter color = more energy.
Whatever your picture is, it isn't a full spectrum for Madonna's 'Vogue'. Everything above 95 Hz is missing. That apparently eliminates the synth strings intro, and I'm guessing the weird gap at 270 sec is the piano/vocal breakdown.
You say you want to know 'where all the energy goes' in a track. You've been told several times now how to determine that: generate a full spectrogram. (I'll save you some time: most of the time, there's more energy in the bass than anywhere else)
But why do you want to know?
