Theory of Everything

bogosort

Joined Sep 24, 2011
696
Can you elaborate? To my knowledge, the tech needed to accomplish what the op has asked does not yet exist.
The "miraculous" part was in reference to the question: "How is the mind able to 'access' or identify that one note and even mentally change just it? " I was pointing out that no one knows is a woefully ignorant answer.
 

crutschow

Joined Mar 14, 2008
34,282
Pychophysics is a mature field, with well over a century's worth of research into how the auditory system works. While there still are some open questions, the nature of sound discrimination is not one of them.
So we understand how the brain (without digital signal processing) can identify all the instruments and vocalists in a recording?
 

MrChips

Joined Oct 2, 2009
30,710
You get the gist of what I’m asking. Thanks for everyone’s reply... the essence I’m getting at is:
The captured recording at any given “time slice” is one parent wave composed of dozens of waves, each of which represents notes being played by all the instruments. So buried at second 3.5 to 4.5 is an A440 of a violin, for example, with all its unique timbre overtones.

How can one get to that one single wave and change it, if, for example, it was “the wrong note?”
Are you asking "How to edit a sound bite electronically?" or "How does the brain pick out a single bad note from the total performance?"

This is an electronics forum and we can answer the former.
With audio editing software we can examine a sound wave as a temporal sequence of voltages. We can identify the annoying wavelet in time, remove it from the background noise and replace it with the correct note and desired timbre.

Since we are not psychoacoustics experts we don't have answers for the latter.
What we do know is that the physical anatomy of the cochlea in the inner ear represents a mechanical spectrum analyzer. It is able to isolate different frequencies and then relay this information to the brain. This is equivalent to what we do in math/engineering/electronics known as the Fourier Transform (named after Jean-Baptiste Joseph Fourier, a French mathematician 1768-1830) which is a method of converting a time series into frequency information.
 

bogosort

Joined Sep 24, 2011
696
So we understand how the brain (without digital signal processing) can identify all the instruments and vocalists in a recording?
Yes. Complex acoustic signals are a linear superposition of simple acoustic signals, which makes them eminently suitable for decomposition, e.g., by a filter bank, such as the human ear. Fundamentally, acoustic information is just modulated harmonic structure. Different parts of our ears respond to the various periodic/harmonic aspects of the modulated signal (tonotopy), which the brain uses to discriminate parts or integrate into a whole. An enormous amount of research has been done revealing the physiological nature and limits of sound discrimination and integration in the human auditory system.
 

MrChips

Joined Oct 2, 2009
30,710
Question. Would that mean then, that nowadays it's possible to take a recording of a trio of voices singing in unison, and with the use of said software one would be able to perfectly and completely isolate or delete one of the voices from the recording?
I would say yes.
They can do it with pictures which is 2-D. A recording is only 1-D.

Edit: In unison, singing the same note? Maybe not. In harmony, yes.
 

nsaspook

Joined Aug 27, 2009
13,081
I would say yes.
They can do it with pictures which is 2-D. A recording is only 1-D.

Edit: In unison, singing the same note? Maybe not. In harmony, yes.
It's unlikely they could synchronize and keep the same note in phase ( phase-locked and phase-coherent ) so each singer would be uncorrelated in at least one important way and individually detectable using ICA.
 

Thread Starter

Jennifer Solomon

Joined Mar 20, 2017
112
I would say yes.
They can do it with pictures which is 2-D. A recording is only 1-D.

Edit: In unison, singing the same note? Maybe not. In harmony, yes.
But in the case of a piano piece with up to 10 notes playing simultaneously, including sustain, reverb, etc — the mind can pick out individual notes and notice a "mistake" with one of them easily. They don't have any near-term software or even theoretics to pull the individual note wavelet's out of the mix, and change it, right?
 

MrAl

Joined Jun 17, 2014
11,389
But in the case of a piano piece with up to 10 notes playing simultaneously, including sustain, reverb, etc — the mind can pick out individual notes and notice a "mistake" with one of them easily. They don't have any near-term software or even theoretics to pull the individual note wavelet's out of the mix, and change it, right?
Not necessarily. The mind has a simpler time when the music is within the content range of the particular human listening experience, but introduce a new piece and the same person may never know the difference between a good note and a bad note.
A simpler example would probably be some Rachmaninoff piano pieces where the music moves fast and does not adhere to any diatonic scale in many cases. Another example some Chopin piano pieces that have chromatic scale waterfall like arpeggios. Many Jazz pieces would have runs that are impossible to understand right from wrong for the untrained ear. An extreme example would be some computer music which i could demonstrate if we can upload music, where it is absolutely impossible for a human to tell if a proper note was played or not.

Can software pick out these individual notes though and change them even if they are right? Well, it depends how well the software can provide a spectrum analysis with the frequencies used in most music.
But then we have the class of music that use those in between frequencies which really have infinite range of tone not just the usually 12 note semi tone scale. The software would also have to be able to read the music score in order to determine if a mistake was made or not, assuming that we place restrictions on the performer where we limit their artistic license to playing the sheet music exactly as written.

So we are dealing with a wide range of possibilities. I would think there would have to be some restrictions on the content in order to determine if any mistakes had been made. I have heard some classical music orchestras play Bach pieces that i could swear they didnt know how to follow very well at all over the entire score:)
 
Last edited:

MrChips

Joined Oct 2, 2009
30,710
But in the case of a piano piece with up to 10 notes playing simultaneously, including sustain, reverb, etc — the mind can pick out individual notes and notice a "mistake" with one of them easily. They don't have any near-term software or even theoretics to pull the individual note wavelet's out of the mix, and change it, right?
Are you asking just out of curiosity or is there an actual problem you would like to solve?
Are you inquiring about the psycho-acoustics of music discrimination or about an automatic or manual note correcting computer based audio editor?

What would be the definition of a "bad" note?
In many performances there are "passing" notes. What would be a "bad" note in jazz improvisation?
With many types of musical instruments, one can "bend" a note over many semi-tones. How would you define a "bad" note in such a context?
 

MrAl

Joined Jun 17, 2014
11,389
Are you asking just out of curiosity or is there an actual problem you would like to solve?
Are you inquiring about the psycho-acoustics of music discrimination or about an automatic or manual note correcting computer based audio editor?

What would be the definition of a "bad" note?
In many performances there are "passing" notes. What would be a "bad" note in jazz improvisation?
With many types of musical instruments, one can "bend" a note over many semi-tones. How would you define a "bad" note in such a context?
That;s exactly what i was saying in my previous post. There has to be restrictions.
 

Thread Starter

Jennifer Solomon

Joined Mar 20, 2017
112
Are you asking just out of curiosity or is there an actual problem you would like to solve?
Are you inquiring about the psycho-acoustics of music discrimination or about an automatic or manual note correcting computer based audio editor?

What would be the definition of a "bad" note?
In many performances there are "passing" notes. What would be a "bad" note in jazz improvisation?
With many types of musical instruments, one can "bend" a note over many semi-tones. How would you define a "bad" note in such a context?
It’s a little of all of it. :) I have a piano performance that has one bad note amidst several playing, and I wonder if there’s a way to “cannibalize” that wave and surgically shift just that one note a whole tone lower. But that’s just one portion of my curiosity concerning the ability to “get at” information embedded in a wave.

I find it very curious that an audio speaker at any given time is vibrating one wave, but yet potentially dozens or hundreds of wavelets composed that parent wave—so that individual parts are distinguishable by the mind—with “air pressure shiftings alone” embedding timbres, tonalities, and other things that make up each individual wavelet’s characteristics, yet somehow all “coalesced” at each point in the parent wave.
 

MrChips

Joined Oct 2, 2009
30,710
It’s a little of all of it. :) I have a piano performance that has one bad note amidst several playing, and I wonder if there’s a way to “cannibalize” that wave and surgically shift just that one note a whole tone lower. But that’s just one portion of my curiosity concerning the ability to “get at” information embedded in a wave.
You can do this manually with the right audio editor. There is no automated editor that will do this that I am aware of.

I find it very curious that an audio speaker at any given time is vibrating one wave, but yet potentially dozens or hundreds of wavelets composed that parent wave—so that individual parts are distinguishable by the mind—with “air pressure shiftings alone” embedding timbres, tonalities, and other things that make up each individual wavelet’s characteristics, yet somehow all “coalesced” at each point in the parent wave.
Yes, you are correct. There is only one wave that reaches the ear. That wave is a collage of a hundred or more different waves combining into one. This is the mystery and beauty of the human brain.

As another example, one can be walking in the woods and observe a flash of colour lasting for no more that a fraction of a second. And yet the brain is able to identify the species of bird that just flew by.
 

Thread Starter

Jennifer Solomon

Joined Mar 20, 2017
112
Jennifer Solomon said:
It’s a little of all of it. :)I have a piano performance that has one bad note amidst several playing, and I wonder if there’s a way to “cannibalize” that wave and surgically shift just that one note a whole tone lower. But that’s just one portion of my curiosity concerning the ability to “get at” information embedded in a wave.
You can do this manually with the right audio editor. There is no automated editor that will do this that I am aware of.
I’ve never heard of one...any link?
 

Thread Starter

Jennifer Solomon

Joined Mar 20, 2017
112
You can do this manually with the right audio editor. There is no automated editor that will do this that I am aware of.


Yes, you are correct. There is only one wave that reaches the ear. That wave is a collage of a hundred or more different waves combining into one. This is the mystery and beauty of the human brain.

As another example, one can be walking in the woods and observe a flash of colour lasting for no more that a fraction of a second. And yet the brain is able to identify the species of bird that just flew by.
The wave would have to be digitized to be stored in the brain, no? Because there is no “continuous medium” in the brain representing the millions received every day? Not only that, but the mind can readily compare the contents of that wave to other waves instantly and effortlessly.

I personally believe the wave is therefore actually not what it appears. The wave is a numeric “key” that is “unlocking” the existing information in the 5D, where infinite other waves can be contrasted, correlated and discretized in their relationship to time and space instantly, but I digress. ;)
 

MrChips

Joined Oct 2, 2009
30,710
The wave would have to be digitized to be stored in the brain, no? Because there is no “continuous medium” in the brain representing the millions received every day? Not only that, but the mind can readily compare the contents of that wave to other waves instantly and effortlessly.

I personally believe the wave is therefore actually not what it appears. The wave is a numeric “key” that is “unlocking” the existing information in the 5D, where infinite other waves can be contrasted, correlated and discretized in their relationship to time and space instantly, but I digress. ;)
We know with a high degree of certainty that what reaches the human ear is an analog pressure wave.
Is it "digitized" and stored in the brain? I doubt that very much.
How the brain processes and stores this information is completely outside of my field of competence.

What I do know is that the brain can recall images, sounds, music, words, phrases, poems, etc. captured decades earlier. For example, the last time I studied and used French vocabulary was 50 years ago. Such words are still in my brain and now returning from my deep memory with ease. My 100-year old mother tells us stories back when she was 15 years old. How the brain is able to do that is a miracle to me.

The experts claim that long term memory happens because of association and recall.
I believe there is more to it than that. I can recall verb conjugation in Latin though I have not used them in intervening years.
 

Thread Starter

Jennifer Solomon

Joined Mar 20, 2017
112
We know with a high degree of certainty that what reaches the human ear is an analog pressure wave.
Is it "digitized" and stored in the brain? I doubt that very much.
How the brain processes and stores this information is completely outside of my field of competence.

What I do know is that the brain can recall images, sounds, music, words, phrases, poems, etc. captured decades earlier. For example, the last time I studied and used French vocabulary was 50 years ago. Such words are still in my brain and now returning from my deep memory with ease. My 100-year old mother tells us stories back when she was 15 years old. How the brain is able to do that is a miracle to me.

The experts claim that long term memory happens because of association and recall.
I believe there is more to it than that. I can recall verb conjugation in Latin though I have not used them in intervening years.
Well, we know it is a wave of infinite points, and it is storing this and millions more waves somehow without an actual geometric representation... sounds entirely like something else is happening that cannot be physically perceived, but that’s just me.

re: editor—I don’t think there is an audio editor that will permit the cannibalization of waves into wavelets for editing... just googled.
 

nsaspook

Joined Aug 27, 2009
13,081
Well, we know it is a wave of infinite points, and it is storing this and millions more waves somehow without an actual geometric representation... sounds entirely like something else is happening that cannot be physically perceived, but that’s just me.
...
Sounds like metaphysical pseudoscience to me.
Obviously it's physically possible as the brain operates within finite physical principles as does the transmission media for the source waves.
 
Top