Silly but fundamental question - kindly illuminate

[sachinchavan 15865]

I always wondered how a speaker (driver) make sounds. No, I am not looking for explanation of the electromagnetic mechanism behind it. Just how the cone makes the sound? I mean when you look at the frequency graph of any track being played, you see that hundreds of (technically infinite number of) frequencies are amplified at any point in time, producing the various vocals, instruments and other artefacts on the track. My question is, how does the seemingly simple, uniform cone produce all these frequencies? Do different parts of it vibrate at different frequencies at the same time (something difficult to visualise). If so, how does that happen in what seems to be a continuous cone? If not, then how are the different sound frequencies produced simultaneously?

I know that the answer, when one of you gives it, might make me look stupid for asking this question. But I had to ask it nevertheless! There might be some other techno-novices like me who have wondered the same sometime and would benefit from your answers.

 

[Yelamanchili manohar]

I always wondered how a speaker (driver) make sounds. No, I am not looking for explanation of the electromagnetic mechanism behind it. Just how the cone makes the sound? I mean when you look at the frequency graph of any track being played, you see that hundreds of (technically infinite number of) frequencies are amplified at any point in time, producing the various vocals, instruments and other artefacts on the track. My question is, how does the seemingly simple, uniform cone produce all these frequencies? Do different parts of it vibrate at different frequencies at the same time (something difficult to visualise). If so, how does that happen in what seems to be a continuous cone? If not, then how are the different sound frequencies produced simultaneously?

I know that the answer, when one of you gives it, might make me look stupid for asking this question. But I had to ask it nevertheless! There might be some other techno-novices like me who have wondered the same sometime and would benefit from your answers.

It is definitely not stupid....I've wondered about the same many times

 

[mbhangui]

1. Any surface that vibrates at a particular frequency makes sound at that frequency.
2. Any surface that vibrates at two or more frequency makes sounds at those frequencies.
3. e.g. a tuning fork of 128 Hz when hit, will vibrate and make sound at 128 Hz
4. Your vocal cords can vibrate at multiple frequencies. Women have tighter/smaller vocal chords compared to men, making them good for higher frequencies compared to men.
5. All objects have a fundamental frequency. Larger, longer objects can vibrate better at lower frequency. Thus long bridges can vibrate at very low frequency. Soldiers marching perfectly on such a bridge can destroy the bridge. Hence when you cross such a bridge, soldiers will be asked not to march in unison. Same way, Larger speaker diagraphm will vibrate better at lower frequencies. Hence woofer are used for bass. Remember "baby ko base pasand hai"? It just means that the babe likes big speakers. Tiny speakers with tight metal diagraphm will vibrate better at higher frequencies. Hence twitters are used for high frequency.
6. You can cut groves on plastic, shellac or any material with a needle and a large heavy metal tin connected to the needle and let the plastic rotate. It will cut groove on the plastic acccording to how the metal tin lid receives sound. When you put needle on the plastic and let the plastic disc rotate, it will produce the same sound. This is where you are storing the sound energy into grooves on the plastic. As an experiment you can just take a needle and let it rest on a record while it is rotating, you will be able to hear the song without any amplifier. Just put your ear near the needle.
7. Instead of mechanically etching the sound, you can store it into other forms, like using a microphone, amplifying it, connect it to a electromagnet over which a magnetic tape slides. This way you willl etch the sound onto a magnetic tape. When you play the same take over any electomagnet, the varying magnetic field will convert it into electricty. If you amplify that electricity and pass it through a coil wound on any damn material that gets attracted to magnet (iron, etc), the material will vibrate and this vibration will produce sound
8. The speaker driver is just a coil wound over a cylinder (usually paper, polymer) that slides over a smaller sized magnet. when you apply electricty to the coil the coil moves. When you apply electrictity at a particular frequency, the paper/polymer cylinder will vibrate at the same frequency. If you connect a paper coil on the coil, it will produce a larger sound. When in 10th standard I just wound multiple rounds of enamelled copper wire on a pan parag tin and connected that to my busted radio speaker terminal and it did the job as a speaker.

There is no rocket science here. I have tried the recording stuff on plastic as a kid (specifically Lakme Vanishing Cream inside plastic lid, recorded voice on it) and did lot of experiments with my grandmothers hand wound gramaphone. There is no magic happening here. This was all discovered and exploited ages ago by people like Faraday, Edison and Tesla. It is just that the medium of storing the sound has changed

 

[arj]

I always wondered how a speaker (driver) make sounds. No, I am not looking for explanation of the electromagnetic mechanism behind it. Just how the cone makes the sound? I mean when you look at the frequency graph of any track being played, you see that hundreds of (technically infinite number of) frequencies are amplified at any point in time, producing the various vocals, instruments and other artefacts on the track. My question is, how does the seemingly simple, uniform cone produce all these frequencies? Do different parts of it vibrate at different frequencies at the same time (something difficult to visualise). If so, how does that happen in what seems to be a continuous cone? If not, then how are the different sound frequencies produced simultaneously?

I know that the answer, when one of you gives it, might make me look stupid for asking this question. But I had to ask it nevertheless! There might be some other techno-novices like me who have wondered the same sometime and would benefit from your answers.

the cone is actually wiggling ie it is a very complex movement. it moves in and out for the lower frequencies and the cone surface moves to the higher frequencies. Easier to imagine an instant of time where there are several frequencies which all get summed up into one but while that is getting reproduced the next instant also needs to be and hence the surface moves rather crazy

( since Even one note from an instrument has several overtones and undertones which all get added up into a summed wave ..just imagine with more instruments how complex it gets)


found this you tube video

 

[sandeepsasi]

Hi Sachin,

It is a very simple question and it is what a kind of makes answering it very difficult (at least for me). I'll try to give an explanation using a simple model at the risk of experts here calling for banning me for oversimplifying concepts.

Let's assume that our chain consists of a DAC driving a power-amp which in turn drives a pair of full-range speakers (so no crossovers to worry about in this case ) and, that the DAC produces a max line level output voltage of 2V at full volume.

Now consider the case when the signal is a pure 1000Hz tone. This is how the output voltage of the DAC looks like, over time:

Note: We are going to consider only one channel in these examples.

We would have come across these kind of waveforms before and nothing really interesting here. Now, consider the case when the signal consists of two tones: a 1000Hz tone and a 10000Hz tone. The first two plots illustrate how the DAC's output voltage, at full volume, vary over time when each tone is applied separately and the third plot illustrates the DAC's output voltage, at full volume, while playing the combined signal.


Note: Here I'm assuming that the amplitudes of the tones have been normalized to get max output voltage at full volume while recording the signal. To put it differently, each tone in this case is assumed to have a peak amplitude of 1V and the combined signal a peak of 1V + 1V = 2V.

We can extend this further and consider a signal obtained by combining 10 tones. At full volume, the plot below illustrates how the DAC's output voltage varies over time:



The tones composing the signal have been mentioned in the plot's title. Would you have believed that 10 sine waves when combined can look so bad?

So far we have only seen how the line level output voltage of the DAC looks like. But, your question is regarding the speaker cone movement. In our example chain, the DAC drives the power-amp and the power-amp drives the speakers. This is where we are going to simplify concepts (a lot!)

Let us assume that a +ve voltage at the DAC's output results in a larger +ve voltage at the power-amp's output which in turn will push the speaker cone forward. Similarly, a -ve voltage at the DAC's output results in a larger -ve voltage at the power-amp's output which pushes the speaker cone backward. When the DAC's output voltage is zero, the power-amp's output voltage is also zero and there is no displacement in the speaker cone (as good as when the chain is powered-off).

By combing electrical, mechanical and acoustic engineering theory and based on complex equations, we may be able to predict (to some extent), what the cone displacement is going to be for a given tone frequency and amplitude. But, we are going to abstract all this using a simple linear equation (experts in these fields can tell how dumb I sound!):

1. A +2V line voltage at the DAC's output is going to displace the speaker cone by, say, +5mm ("+" indicates forward movement of the cone)
2. A -2V line voltage at the DAC's output is going to displace the speaker cone by, say, -5mm ("-" indicates backward movement of the cone)
3. 0V at the DAC's output is going to have no effect on the speaker cone



Note: I've taken the liberty (or atrocity?) to ignore the tone frequency in this model!

Now, it is just a matter of changing the scale and units of the above graphs, to get plots of cone displacement in mm over time, for a given line level voltage waveform.

First, the 1000Hz tone's case. The first plot below shows the DAC's output voltage over time and the second plot shows the speaker cone displacement in mm over time (based on our unrealistic, oversimplified, linear model, of course ):



Now, let's look at the case of the mult-tone signal.


In this case, you can see that the cone will have to often travel a lot of distance, forward and backward, very quickly. Practically, it is very difficult for a single driver to accomplish this. This is loosely why full range drivers have a limited bandwidth (usually limited at extreme ends of the audio spectrum) and we often choose to go for multi-driver speakers, as each driver can be optimized to handle only a certain range of frequencies.

Hope this was helpful!

With regards,
Sandeep Sasi

 

[sachinchavan 15865]

the cone is actually wiggling ie it is a very complex movement. it moves in and out for the lower frequencies and the cone surface moves to the higher frequencies. Easier to imagine an instant of time where there are several frequencies which all get summed up into one but while that is getting reproduced the next instant also needs to be and hence the surface moves rather crazy

( since Even one note from an instrument has several overtones and undertones which all get added up into a summed wave ..just imagine with more instruments how complex it gets)


found this you tube video

Thanks @arj. The video explains it to an extent. Also what you said - the cone moving in and out for lower frequencies and cone surface moving to higher frequencies was interesting addition to my knowledge. However, still an important question remains unaddressed in my mind. Asking that in the end of this reply.

1. Any surface that vibrates at a particular frequency makes sound at that frequency.
2. Any surface that vibrates at two or more frequency makes sounds at those frequencies.
3. e.g. a tuning fork of 128 Hz when hit, will vibrate and make sound at 128 Hz
4. Your vocal cords can vibrate at multiple frequencies. Women have tighter/smaller vocal chords compared to men, making them good for higher frequencies compared to men.
5. All objects have a fundamental frequency. Larger, longer objects can vibrate better at lower frequency. Thus long bridges can vibrate at very low frequency. Soldiers marching perfectly on such a bridge can destroy the bridge. Hence when you cross such a bridge, soldiers will be asked not to march in unison. Same way, Larger speaker diagraphm will vibrate better at lower frequencies. Hence woofer are used for bass. Remember "baby ko base pasand hai"? It just means that the babe likes big speakers. Tiny speakers with tight metal diagraphm will vibrate better at higher frequencies. Hence twitters are used for high frequency.
6. You can cut groves on plastic, shellac or any material with a needle and a large heavy metal tin connected to the needle and let the plastic rotate. It will cut groove on the plastic acccording to how the metal tin lid receives sound. When you put needle on the plastic and let the plastic disc rotate, it will produce the same sound. This is where you are storing the sound energy into grooves on the plastic. As an experiment you can just take a needle and let it rest on a record while it is rotating, you will be able to hear the song without any amplifier. Just put your ear near the needle.
7. Instead of mechanically etching the sound, you can store it into other forms, like using a microphone, amplifying it, connect it to a electromagnet over which a magnetic tape slides. This way you willl etch the sound onto a magnetic tape. When you play the same take over any electomagnet, the varying magnetic field will convert it into electricty. If you amplify that electricity and pass it through a coil wound on any damn material that gets attracted to magnet (iron, etc), the material will vibrate and this vibration will produce sound
8. The speaker driver is just a coil wound over a cylinder (usually paper, polymer) that slides over a smaller sized magnet. when you apply electricty to the coil the coil moves. When you apply electrictity at a particular frequency, the paper/polymer cylinder will vibrate at the same frequency. If you connect a paper coil on the coil, it will produce a larger sound. When in 10th standard I just wound multiple rounds of enamelled copper wire on a pan parag tin and connected that to my busted radio speaker terminal and it did the job as a speaker.

There is no rocket science here. I have tried the recording stuff on plastic as a kid (specifically Lakme Vanishing Cream inside plastic lid, recorded voice on it) and did lot of experiments with my grandmothers hand wound gramaphone. There is no magic happening here. This was all discovered and exploited ages ago by people like Faraday, Edison and Tesla. It is just that the medium of storing the sound has changed

@mbhangui, appreciate the effort taken in elaborately describing the concept behind recording and playback. I knew most of it though, still your comprehensive explanation is very educational. However, I am specifically looking to understand the last stage - how the cone’s movements reproduce complex music? Trying to put it together from the replies from you all.

Hi Sachin,

It is a very simple question and it is what a kind of makes answering it very difficult (at least for me). I'll try to give an explanation using a simple model at the risk of experts here calling for banning me for oversimplifying concepts.

Let's assume that our chain consists of a DAC driving a power-amp which in turn drives a pair of full-range speakers (so no crossovers to worry about in this case ) and, that the DAC produces a max line level output voltage of 2V at full volume.

Now consider the case when the signal is a pure 1000Hz tone. This is how the output voltage of the DAC looks like, over time:
View attachment 71568
Note: We are going to consider only one channel in these examples.

We would have come across these kind of waveforms before and nothing really interesting here. Now, consider the case when the signal consists of two tones: a 1000Hz tone and a 10000Hz tone. The first two plots illustrate how the DAC's output voltage, at full volume, vary over time when each tone is applied separately and the third plot illustrates the DAC's output voltage, at full volume, while playing the combined signal.

View attachment 71569
Note: Here I'm assuming that the amplitudes of the tones have been normalized to get max output voltage at full volume while recording the signal. To put it differently, each tone in this case is assumed to have a peak amplitude of 1V and the combined signal a peak of 1V + 1V = 2V.

We can extend this further and consider a signal obtained by combining 10 tones. At full volume, the plot below illustrates how the DAC's output voltage varies over time:

View attachment 71570

The tones composing the signal have been mentioned in the plot's title. Would you have believed that 10 sine waves when combined can look so bad?

So far we have only seen how the line level output voltage of the DAC looks like. But, your question is regarding the speaker cone movement. In our example chain, the DAC drives the power-amp and the power-amp drives the speakers. This is where we are going to simplify concepts (a lot!)

Let us assume that a +ve voltage at the DAC's output results in a larger +ve voltage at the power-amp's output which in turn will push the speaker cone forward. Similarly, a -ve voltage at the DAC's output results in a larger -ve voltage at the power-amp's output which pushes the speaker cone backward. When the DAC's output voltage is zero, the power-amp's output voltage is also zero and there is no displacement in the speaker cone (as good as when the chain is powered-off).

By combing electrical, mechanical and acoustic engineering theory and based on complex equations, we may be able to predict (to some extent), what the cone displacement is going to be for a given tone frequency and amplitude. But, we are going to abstract all this using a simple linear equation (experts in these fields can tell how dumb I sound!):

1. A +2V line voltage at the DAC's output is going to displace the speaker cone by, say, +5mm ("+" indicates forward movement of the cone)
2. A -2V line voltage at the DAC's output is going to displace the speaker cone by, say, -5mm ("-" indicates backward movement of the cone)
3. 0V at the DAC's output is going to have no effect on the speaker cone

View attachment 71573

Note: I've taken the liberty (or atrocity?) to ignore the tone frequency in this model!

Now, it is just a matter of changing the scale and units of the above graphs, to get plots of cone displacement in mm over time, for a given line level voltage waveform.

First, the 1000Hz tone's case. The first plot below shows the DAC's output voltage over time and the second plot shows the speaker cone displacement in mm over time (based on our unrealistic, oversimplified, linear model, of course ):

View attachment 71574

Now, let's look at the case of the mult-tone signal.

View attachment 71572
In this case, you can see that the cone will have to often travel a lot of distance, forward and backward, very quickly. Practically, it is very difficult for a single driver to accomplish this. This is loosely why full range drivers have a limited bandwidth (usually limited at extreme ends of the audio spectrum) and we often choose to go for multi-driver speakers, as each driver can be optimized to handle only a certain range of frequencies.

Hope this was helpful!

With regards,
Sandeep Sasi

@sandeepsasi, many thanks for explaining in such detail what happens with the sound waves. I won’t say I grasped it entirely, but kind of got the gist.

So here’s the question in my mind after reading your explanation as well as the video Arjun posted. As I understand now, the various frequencies (waves) representing each instrument in the music (as well as their harmonics) all combine into a complex superimposed wave. Ok so far. And this wave is what the cone responds to by moving itself. Now, the big question. If all the instruments have combined into this single superimposed wave that reaches our ears, how is it that our brain is able to analyse it (break it up) back into the individual waves and distinctly make out all the instruments that were playing and together made this complex wave that fell on our ears?

 

[arj]

So here’s the question in my mind after reading your explanation as well as the video Arjun posted. As I understand now, the various frequencies (waves) representing each instrument in the music (as well as their harmonics) all combine into a complex superimposed wave. Ok so far. And this wave is what the cone responds to by moving itself. Now, the big question. If all the instruments have combined into this single superimposed wave that reaches our ears, how is it that our brain is able to analyse it (break it up) back into the individual waves and distinctly make out all the instruments that were playing and together made this complex wave that fell on our ears?

hmm..Good food for thought !
Another perspective, thats what happens when we listen to a voice as well ie the sound coming out of our throat is also that one complex wave with so many nuances which the ear can decipher..
eg if we take the complex/combined multi tone which @sandeepsasi mentioned

Perhaps one perspective could be to separate that into time domain and frequency domain . ie at one instant we hear only 1 freq which is the combination of all waves in that instant but when all these "instants" are combined we hear the spectrum ?

Our ear is actually putting all of this in the time domain and also doing some filtering in terms of removing unnecessary room reflections etc to for the signals to the brain,

 

[keith_correa]

If all the instruments have combined into this single superimposed wave that reaches our ears, how is it that our brain is able to analyse it (break it up) back into the individual waves and distinctly make out all the instruments that were playing and together made this complex wave that fell on our ears?

IMO, we are overrating the role of the brain in this and as a consequence, underrating the role of the ear and related organs. AFAIK, it is the ear et al that breaks the sound down and the brain just post processes it. The ear functions as a sort of very advanced Fourier analysis device. Look it up - it's makes for fascinating reading and is not an exact science according to some, who say that the science is still evolving on this.

 

[arj]

IMO, we are overrating the role of the brain in this and as a consequence, underrating the role of the ear and related organs. AFAIK, it is the ear et al that breaks the sound down and the brain just post processes it. The ear functions as a sort of very advanced Fourier analysis device. Look it up - it's makes for fascinating reading and is not an exact science according to some, who say that the science is still evolving on this.

True that ie in bold. the shape of our ear actually helps in height perception of sound which includes the external ear cartilage and the internal acoustics. all of that helps in collection and perception of sound and converts into electrical signals.

Not sure of how much or little the brain does. the way i understood the filteration and convertson is done by the ear but the brain is the only one which processes/discards/amplifies and in some cases reconstructs the signals

we have some medical folks here may be they can educate us... @Analogous ?

 

[keith_correa]

IMO, we are overrating the role of the brain in this and as a consequence, underrating the role of the ear and related organs. AFAIK, it is the ear et al that breaks the sound down and the brain just post processes it. The ear functions as a sort of very advanced Fourier analysis device. Look it up - it's makes for fascinating reading and is not an exact science according to some, who say that the science is still evolving on this.

I think I sort of underrated the role of the brain by using the word "just". I apologize!

 

[Hari Iyer]

The speakers have been around here for more than 120+ years and this technology is relatively the same and has not yet been challenged. There might be some improvements in motor system, magnets and cone material but the basic science is still the same.

 

[sachinchavan 15865]

True that ie in bold. the shape of our ear actually helps in height perception of sound which includes the external ear cartilage and the internal acoustics. all of that helps in collection and perception of sound and converts into electrical signals.

Not sure of how much or little the brain does. the way i understood the filteration and convertson is done by the ear but the brain is the only one which processes/discards/amplifies and in some cases reconstructs the signals

we have some medical folks here may be they can educate us... @Analogous ?

Ok, so the question now turns to:

“How does the ear convert the complex multi-tone wave it receives from the driver cone back into its constituents (instruments, voices etc)?”

 

[arj]

Ok, so the question now turns to:

“How does the ear convert the complex multi-tone wave it receives from the driver cone back into its constituents (instruments, voices etc)?”

Since in the end at that slice of time ie that instant, it is just one frequency.

Our ear drum is like a reverse speaker which converts it back to an electric signal which the brain decodes. voice/instrument etc is as to how our brain recognises the signal based on our memory

I am sure any animal say a Rat ,even with a similiar ear structure, may not know to differentiate into instrument/voices..that will just be sound to it. ( my assumption)

I guess we are now moving away from audio to the brains audio cognitory abilities so definitely out of my zone of understanding !

 

[dvraghavan]

Hi Sachin,

It is a very simple question and it is what a kind of makes answering it very difficult (at least for me). I'll try to give an explanation using a simple model at the risk of experts here calling for banning me for oversimplifying concepts.

Let's assume that our chain consists of a DAC driving a power-amp which in turn drives a pair of full-range speakers (so no crossovers to worry about in this case ) and, that the DAC produces a max line level output voltage of 2V at full volume.

Now consider the case when the signal is a pure 1000Hz tone. This is how the output voltage of the DAC looks like, over time:
View attachment 71568
Note: We are going to consider only one channel in these examples.

We would have come across these kind of waveforms before and nothing really interesting here. Now, consider the case when the signal consists of two tones: a 1000Hz tone and a 10000Hz tone. The first two plots illustrate how the DAC's output voltage, at full volume, vary over time when each tone is applied separately and the third plot illustrates the DAC's output voltage, at full volume, while playing the combined signal.

View attachment 71569
Note: Here I'm assuming that the amplitudes of the tones have been normalized to get max output voltage at full volume while recording the signal. To put it differently, each tone in this case is assumed to have a peak amplitude of 1V and the combined signal a peak of 1V + 1V = 2V.

We can extend this further and consider a signal obtained by combining 10 tones. At full volume, the plot below illustrates how the DAC's output voltage varies over time:

View attachment 71570

The tones composing the signal have been mentioned in the plot's title. Would you have believed that 10 sine waves when combined can look so bad?

So far we have only seen how the line level output voltage of the DAC looks like. But, your question is regarding the speaker cone movement. In our example chain, the DAC drives the power-amp and the power-amp drives the speakers. This is where we are going to simplify concepts (a lot!)

Let us assume that a +ve voltage at the DAC's output results in a larger +ve voltage at the power-amp's output which in turn will push the speaker cone forward. Similarly, a -ve voltage at the DAC's output results in a larger -ve voltage at the power-amp's output which pushes the speaker cone backward. When the DAC's output voltage is zero, the power-amp's output voltage is also zero and there is no displacement in the speaker cone (as good as when the chain is powered-off).

By combing electrical, mechanical and acoustic engineering theory and based on complex equations, we may be able to predict (to some extent), what the cone displacement is going to be for a given tone frequency and amplitude. But, we are going to abstract all this using a simple linear equation (experts in these fields can tell how dumb I sound!):

1. A +2V line voltage at the DAC's output is going to displace the speaker cone by, say, +5mm ("+" indicates forward movement of the cone)
2. A -2V line voltage at the DAC's output is going to displace the speaker cone by, say, -5mm ("-" indicates backward movement of the cone)
3. 0V at the DAC's output is going to have no effect on the speaker cone

View attachment 71573

Note: I've taken the liberty (or atrocity?) to ignore the tone frequency in this model!

Now, it is just a matter of changing the scale and units of the above graphs, to get plots of cone displacement in mm over time, for a given line level voltage waveform.

First, the 1000Hz tone's case. The first plot below shows the DAC's output voltage over time and the second plot shows the speaker cone displacement in mm over time (based on our unrealistic, oversimplified, linear model, of course ):

View attachment 71574

Now, let's look at the case of the mult-tone signal.

View attachment 71572
In this case, you can see that the cone will have to often travel a lot of distance, forward and backward, very quickly. Practically, it is very difficult for a single driver to accomplish this. This is loosely why full range drivers have a limited bandwidth (usually limited at extreme ends of the audio spectrum) and we often choose to go for multi-driver speakers, as each driver can be optimized to handle only a certain range of frequencies.

Hope this was helpful!

With regards,
Sandeep Sasi

Very detailed explanation. Felt like reading Fourier analysis from my engineering days.

 

[jls001]

Ok, so the question now turns to:

“How does the ear convert the complex multi-tone wave it receives from the driver cone back into its constituents (instruments, voices etc)?”

All sounds are complex multi tones. Pure tones of a single frequency probably exists only from electronic test signal sources. Even synthesized instrument sounds have a set of harmonics which imparts it its uniqueness (besides the modulation - sawtooth, square, sine, etc).

I would speculate that the brain identifies a set of harmonics and identifies that as so and so sound. Of course I could be totally wrong

 

[sachinchavan 15865]

Since in the end at that slice of time ie that instant, it is just one frequency.

Yes, that’s what baffles me. How does the ear/brain/ear-brain combo resolve it back into the constituent frequencies that combined together? This is what this exploration is narrowing down to.

Our ear drum is like a reverse speaker which converts it back to an electric signal which the brain decodes. voice/instrument etc is as to how our brain recognises the signal based on our memory

Yes, I understand that prior familiarity with the instruments/constituents is essential for the brain to identify/decode the instruments/constituents. Though that too is fascinating, am not currently interested in it, but the above (in bold).

 

[arj]

Yes, that’s what baffles me. How does the ear/brain/ear-brain combo resolve it back into the constituent frequencies that combined together? This is what this exploration is narrowing down to.

I can be accused of Conjecture here...but I guess its when the overall signal is made that the brain recognises it as a sequence.

ie the ear takes a slice of time of a freq converts it as a electric signal. Since this is a continuous process all of these slices go as one continuous signal to the brain . The brain has filters to remove Noise and then perhaps a pattern recognition based on experience and Natural intelligence algorithms to put it together as instruments. in any instant its one freq but when you integrate it across it becomes the full music.

This perhaps does also explain timing issues in music where due to some reason some frequencies get messed up in the time domain before reaching the ear and hence you brain feels there is something off.

 

[sachinchavan 15865]

I can be accused of Conjecture here...but I guess its when the overall signal is made that the brain recognises it as a sequence.

ie the ear takes a slice of time of a freq converts it as a electric signal. Since this is a continuous process all of these slices go as one continuous signal to the brain . The brain has filters to remove Noise and then perhaps a pattern recognition based on experience and Natural intelligence algorithms to put it together as instruments. in any instant its one freq but when you integrate it across it becomes the full music.

This perhaps does also explain timing issues in music where due to some reason some frequencies get messed up in the time domain before reaching the ear and hence you brain feels there is something off.

Is it why the part of the brain that processes the audio signal is recieves also called ‘temporal’ lobe? Temporal = to do with time. The brain perhaps, as you conjecture, senses/interprets the audio not at any point in time (like say a meter would measure the frequency), but through the sequence over time (even if milliseconds)? Let’s hope @Analogous and other medicos and medical scientists in the forum enlighten us on this.

Having said that about the ‘meter’ above, it made me think further. The app on my mobile - how does it give me a frequency graph (at any moment) if the cone has combined all the tones into a complex wave (which should have only one frequency at any instant)?

Or do we get a clue here? The mobile’s mic and other related hardware and software would also have least count when it comes to time… so what we see as frequency graph is the sum total of the frequencies of the unified wave during that least count time? And if so, could something similar be happening within the least count that our hardware and software (the brain) has?

 

[sachinchavan 15865]

Is it why the part of the brain that processes the audio signal is recieves also called ‘temporal’ lobe? Temporal = to do with time. The brain perhaps, as you conjecture, senses/interprets the audio not at any point in time (like say a meter would measure the frequency), but through the sequence over time (even if milliseconds)? Let’s hope @Analogous and other medicos and medical scientists in the forum enlighten us on this.

Having said that about the ‘meter’ above, it made me think further. The app on my mobile - how does it give me a frequency graph (at any moment) if the cone has combined all the tones into a complex wave (which should have only one frequency at any instant)?

Or do we get a clue here? The mobile’s mic and other related hardware and software would also have least count when it comes to time… so what we see as frequency graph is the sum total of the frequencies of the unified wave during that least count time? And if so, could something similar be happening within the least count that our hardware and software (the brain) has?

Further to my comment above (since it’s a build up on it, posting as a follow-up comment… mods please excuse the transgression).

I came across this interesting tidbit on Wikipedia while constructing my reply on timbre in another thread. It’s about the sound envelope. I sense we might be upto something with the temporality hypothesis above in understanding how the brain resolves a single complex wave into the constituents. I quote from Wikipedia below:

“The physical characteristics of sound that determine the perception of timbre include frequency spectrum and envelope.” And,

“In sound and music, an envelope describes how a sound changes over time. It may relate to elements such as amplitude (volume), frequencies (with the use of filters) or pitch.”

Yes, this is about perception of production and perception of timbre… the envelope (temporality) influences it significantly. Perhaps this phenomenon is not just limited to perception of just the timbre, but even other aspects of the sound received by our brain? Is there a clue in it for what we are trying to understand/decode?

 

[arj]

Is it why the part of the brain that processes the audio signal is recieves also called ‘temporal’ lobe? Temporal = to do with time. The brain perhaps, as you conjecture, senses/interprets the audio not at any point in time (like say a meter would measure the frequency), but through the sequence over time (even if milliseconds)? Let’s hope @Analogous and other medicos and medical scientists in the forum enlighten us on this.

I remember some metrics while I was researching acoustics.. apparently sound sticks in our ear for .1 sec hence if you want to hear an echo of a wave, it needs to come in after .1s. also if any similiar sound wave reaches the ear after 5ms (?) it ignores it as a reflected wave of the same sound it has already received. I guess there are events like Doppler shifts also in play.

Having said that about the ‘meter’ above, it made me think further. The app on my mobile - how does it give me a frequency graph (at any moment) if the cone has combined all the tones into a complex wave (which should have only one frequency at any instant)?

Or do we get a clue here? The mobile’s mic and other related hardware and software would also have least count when it comes to time… so what we see as frequency graph is the sum total of the frequencies of the unified wave during that least count time? And if so, could something similar be happening within the least count that our hardware and software (the brain) has?

I would assume the the app can either give instantaneous or consolidated values In a mobile app even the definition of instantaneous would be some sort if consolidation eg 1s else we may not be able to make sense.

In professional equipment it might be a different case and should get the above

I have used an app called Spectroid which gives an almost instantaneous output and also keeps the maximum across the timeframe as a reference curve above,