20
апр
Audio ear training for recording musicians and sound engineers. Get audio ear training online, improve core listening skills like frequency.
I like the idea but why should I train? The ability to connect what is in your mind with the appropriate parameters you have to dial to get that sound is not an easy task. The steps involved should be: Sometimes people get lost in the translation step and start turning knobs without confidence. The more you work, the better you understand what those knobs really do, but it is a slow process.
People excel in this matter after many years, because they have learned experimenting with lots of different processes applied to lots of different sources. The purpose of this training is to open your ears to what each frequency sounds like and reduce the amount of time needed to acquire this knowledge. In 15 minutes you can guess or correct 100 random equalisations, so training every day for a few weeks is equivalent to accumulating the experience of many years. Where does this training method come from?
Is the Golden Ears Audio Training Program any good? I've heard good reviews from people that used it but Im not too sure. I seen it on Ebay for $50 so I wanted to know do you think it's worth it?
Did you make it up? No, we didn’t! The Guess Method has been used for over 40 years. There were exercises on cassettes, CD’s or it was done manually by teachers. We know it works and it has proven its efficiency over the years with thousands of successful engineers. The new Correct Method comes from a suggestion (the mastering guru) made to us: I suggest presenting a piece of recorded music which needs 'help' in some frequency range and see if the student can not only identify the problem, but also correct for it in a musically satisfying way.
So that is what we did: correct instead of guess. We tested it and we loved it. It turned out to be both funny and instructive. Both methods complement each other really well. Besides, we have designed the software to turn these methods into a limitless tool, way better than any cassette or CD, because you can feed it with any music or sound and configure the exercises to a level of diversity and detail which wasn’t imaginable before. We can’t place more emphasis on this way of training the ear: IT WORKS!
Philips: Golden Ears Training A few months ago when Mike and I were at the in Belgium together with Tyll and Jude, Philips showed us their internal “Golden Ears” testing tool/program. I remember us all being very enthusiast about it, recommending them to make it available to the big public. And guess what? “We know that we are not alone in this obsession with sound, which is why we are inviting you to take our Golden Ears challenge”. The Golden Ears website has been online for a few months now even though Philips hasn’t really been marketing it.
That shows when you look at the number of finished tests on the website but I’m sure these numbers will increase quickly. So what is it exactly?
The golden Ears training program was designed to develop the listening skills of their audio engineers. Their expertise guarantees the creation of products of superior sound quality that highlight every musical detail.
Graduates of the program can reliably detect subtle differences in sound based on 5 key attributes: Timbre: Differences in tone or the frequency of content of music Details: The lack of treble, distortion, or noise that can mask the fine details Spatial Impression: The depth and spaciousness of the sound field Bass: The quality of low frequencies Loudness: The perception of sound power, from quiet to loud This online training program contains elements of their internal program we tried at their facility. To be fair, I do think their internal program is harder to complete than this online training program, but that seems only normal. This online version has four big levels, with short challenges in each of the above mentioned categories: Basic Level – Bronze Ears – Silver Ears – Golden Ears The thing I really liked about the training is that you get to experience a lot of the terms we use to describe sound in our reviews.
In example, if you sometimes wonder what a small sound stage or treble cut off sounds like, this is the place to learn. How long you will take to work your way to the end of the Golden Ears status is hard to predict but with an above average experience in sound like most of our readers have and a good dac/amp/headphone setup, you will (hopefully) go through it fairly quick. Don’t be mistaken however, there are some hard challenges in the Golden part for sure, especially the Timbre section.
Of course you have to start with the Basic level before you can go to the next level but you can do Loudness first and then go to Timbre and mix things up. If you answer any question incorrectly, you will need to go back one step. For your reference, there are 7 chapters in Basic Level; 6 chapters in Bronze Ears; 7 chapters in Silver Ears; 6 chapters and a quiz in the ultimate level – Golden Ears. After you’ve created a log in, the site will remember where exactly you were in the training so you can come back at any given time and continue where you quit. After completing a level you can share that with your friends on the social media. Who wouldn’t want to show off his Golden Ears status with his (audiophile) friends? My non audio friends of course now think of me as an even bigger freak, but what do they know, right?
I sincerely had fun doing all the tests using 2 different setups depending on my location: Cypher Labs Theorem DAC + Beyerdynamic A20 + Sennheiser HD650 and Meier Audio Corda Classic + Daccord + Hifiman HE-500. Philips recommends using the best headphones/setup you have available to complete the program. You can go have a look at, or even better, participate in the training program right here: Let us know if you have made it to the end and what you considered the hardest test was. To me the Timbre section in Gold was very tough. I hope you have as much fun as I had.
Thank you Philips for making this available to the big crowd, for free! (A mobile version of the site is coming soon) adrotate banner=”48″. If you see what I said about the colorations in the headphone you’re doing the tests with, and you’ve noted what Innerfidelity and others have said – that even flagships vary by up to plus or minus 5 db – then how do you propose to evaluate colorations that are smaller than your own headphone’s coloration? I’m going to guess in advance that you’re confident you can do that, and maybe you know a trick that works reliably.
But outside of using a Harman K812 or Sennheiser 800, I doubt the very premise when the differences in the test examples are small enough. And BTW, if the test really takes 10 times longer than I spent (5 sections times 2 – beginning and advanced), who is going to have that much free time? Experts I’d guess. I have to say I flew through everything except for Timbre on gold. Of course good gear is important.
I mentioned the gear I used in the article. The test might not be perfect and yes headphones can have variations, but if you know your headphone, you know what to listen to and you shouldn’t have to much problems IF your ears are good enough. That may seem like a weird statement but I have friends who can’t even make out a wider sound stage, some people just can’t hear those things so I suppose it will be harder for those trainees. I don’t know how Philips is generating the colorations, but based on my recent experience with Audioforge and how it can improve soundstage and eliminate the ‘constricted’ sound in closed headphones, I’d say they can generate those colorations with freq. Response only. And if that’s true, then a very neutral headphone would be needed.
The only reason I can think of that people can use very colored headphones in the training is when they’re lucky enough that a coloration in the headphone doesn’t coincide with a coloration in the test sample. Or, using the example from wave coincidence and reinforcement, that the headphone would cancel one real test track example and color another example that’s not colored. You know how it is that people are born seeing upside down and the brain has to reverse the picture?
That’s a potential problem with this training – you’ll train your brain to hear critically with a very colored headphone, which isn’t a good idea I don’t think. You should start with frequency response. Not only do many pricy headphones have peaks and recesses that you have to somehow compensate for in listening or judging, but frequency response deviations have a major impact on soundstage and other things, especially from the upper mids through the treble. One problem I see in Philips’ tests is the female voice, where the colorations they demonstrate involve frequency response deviations that are less than most headphones, then to make matters worse, the fundamentals where those deviations occur are often 10 to 20 db below the average signal. There are very brief moments where the fundamentals rise in volume, but I don’t know if it’s enough, especially given the colorations in our headphones. Another problem is the absolute secrecy as L said – we won’t know how any experts or reference people did on the tests, except to take their word for it (which I’d love to hear, but can’t make book on).
Generally, I’m trying to get feedback on my question mainly as something funthink of it as a hypothetical question, such as, what would I do if I won a million bucks:). I understand that you may not use scores or preset criteria, but was hoping to read some feedback as if you were given a class assignment to list 5 attributes you find most important to describe the sound quality of a product. I have an idea of what I would use.
Mainly I’m asking this because as a novice reviewer, I make short video reviews. Written reviews can literally be pages upon pages to describe the smallest nuances of how something sounds, which gives the writer a lot of real estate to review something in deep detail.
But, If I were to do a quick ‘nugget’ review in which I provide approximately 5 key attributes to quickly describe the most common aspects of what something sounds like, what would they be? This is of course a more subjective question, hence why I’m seeking more input. Also, I understand that people who review speakers, will perhaps key in on different attributes when compared to folks who generally only review IEM’s or full sized cans. My hearing is perfect.
Ear Training For Audio Engineers
But when someone is trying to whisper in my right ear and someone else is shouting in my left ear, I usually don’t hear the person on the right – not very clearly anyway. The way you claim you can hear perfectly through all of the colorations of your headphone to clearly hear the colorations in the samples – I don’t buy it.
It’s unscientific and illogical. Some you can hear, yes, but not all. Think about that some more and then if you have a logical explanation I’m sure interested. It just means that when two things are playing at the same time, there are cases where one sound “masks” the other sound, unless the 2 sounds are very different from each other and not too far apart in volume. In regard to the Philips training, if the headphone has enough coloration (i.e. The “really crappy” headphone one user noted), some of those colorations would coincide with colorations in the test music, and then correctly identifying the colored music sample would be just a guess. It’s like using a broken calculator that generates occasional wrong numbers when you’re building a bridge and making measurements.
I’m not sure if I understand you correctly, since english is not my first language, but all I’m saying is that I completed the test on gold difficulty with above mentioned headphones. But I never said that it was easy to do so – 2nd part of timbre test on gold took me like 2 hours. And I really don’t know what kind of “logical explanation” are you expecting from me? I simply passed the test and that’s all, but I can post screenshot from it + foto of my pair of HD 600’s / etys hf5 if it helps you somehow. The problem is, this is a new challenge for me, but much more important than that, it’s a major initiative that’s going to play a big role in defining the preferred sound of flagship (or lower) headphones going forward.
It’s not just Philips – Harman is working on this, and so is that pesky guy Paul Barton. So my interest is in finding exactly how people are hearing through one coloration to hear another, especially when (in some cases) it’s a similar coloration and one is going to mask the other. I’m not trying to give anyone a hard time, but we really need to understand some of the basics here instead of just “Hey – I heard it, well, maybe I did”. There is a training aspect to this, but training with very imperfect tools seems at least a little dubious to me.
I don’t buy it. And you can’t say something is plainly wrong just because Innerfidelity “proved” it wrong or right. What you can do IF you want to is demonstrate how you can have a headphone coloration that’s essentially the same as the coloration in the test track and “listen through” one to hear the other. You can’t demonstrate that by saying “I told you so”. And I don’t consider Innerfidelity to be any more than a testing lab. I find their measurements reasonably accurate and useful, but I don’t value their opinions as to what they hear – they have to stand in that line like everyone else.
And it’s a long line. I said it was wrong because I tried it on myself – do I have Stax or other expensive headphones? I have only HD 600 and etys hf5 with odac+o2.
Did I completed the test? Don’t know how for you, but for me it’s clear evidence that one don’t need super expensive headphones to complete that test.
Why it is so? I don’t know and to be honest I don’t really care, but just my 2 cents – you don’t need to care about headphones own coloration, you only need to focus for difference between two provided samples since that hp coloration is present in both samples. Sure there are exception to this, like bass/treble extension. Btw, I just finished bronze with super crappy Samsung earphones, going to try silver. Take the example of a headphone with output at 4 and 6 khz that’s -3 db compared to 5 khz. Then take a ‘colored’ music track that’s had its sound increased by 3 db at 4 and 6 khz.
Now that track will cancel out with the colored headphone and the uncolored track will exhibit the coloration from the headphone. Lemon tree fool's garden meaning. So in that particular case, what does training do? Teach us to hear upside down because we have a colored headphone?
In the real world, those examples I mention here don’t generally occur with the mathematical exactness I stated, but the principle is the same – the colored headphone will make some of the colored music sound better than the uncolored music, and that fact is proven in a hundred places. Beats is a large example, but there are many others. Philips isn’t doing this as a public service – this is going to feed into their ongoing designs, and that’s what I want to know more about – i.e., what’s the real goal for Philips? Well, exactly. Let’s just mark such a headphones as a “Headphone1”, uncolored music sample as A, colored sample as B. Wth are you talking about?
Golden Ears Audio Ear Training
How this deny EQ at all? You should either read again what I wrote or just stop trolling. And I was trying to explain it to you in numerous posts, but no matter what, you are still repeating the same thing that one can’t hear subtle differences because of headphone coloration, but without providing any real evidence to support your theory.
Said, his friend completed that test on gold with koss porta pro, and I did the same with hd600/hf5/samsung mobile earphones. I think you are just afraid to admit, that your hearing is not that good as you thought. It is pointless when you don’t respond to the point. So far, nobody has addressed the specific question as to how to hear a coloration that has been effectively canceled by an opposite coloration in the headphone. I suppose there’s an assumption that it would ‘rarely’ happen, but that’s a potential problem. In the early days of personal computers, virus scans, and a few other things, some of the ‘smart’ programs would look for certain byte signatures in target data to initiate an action, and if the particular signature were statistically very unlikely to impossible to happen otherwise, the programs would execute their action merrily without a second thought.
The WW2 Germans were confident of their Enigma machine’s security based on their statistics – wrong again! So today the success of certain people using the Philips training with very colored headphones has created a new wave of confidence that the headphone doesn’t matter, but I know it does matter, and so should you. Just take a very simple example of a headphone with a 5 db peak at 5 khz, then the test sample has a dip at 5 khz. They will cancel each other out. People can imagine all kinds of things, and even learn workarounds for some of these test things, but they can’t invent science. Waves reinforce and cancel, which is why forcing stereo to mono loses actual sounds, and why headphone crossfeed has made very little progress. To the extent that this training works at all, it’s because the colorations don’t coincide exactly.
But it isn’t good training to learn to detect colorations with a very colored headphone. Philips didn’t do this as a public service. They want to sell headphones, and not the colored kind.
Music Ear Training Free
The actual question is, how does one “train” to hear colorations that are heard with colored headphones, if the colorations in the music and the headphone coincide (either reinforce or cancel out, per ordinary wave physics)? 100 people can post their satisfactory experiences here, but we aren’t learning anything from that until someone does a critical analysis which explains it.
I know that there are forums where people relate their experiences through 100 to 1000 pages of postings that few people will ever read, but I’ve been hoping that on this excellent forum someone would dig into this and provide some real information. I completed the golden ears with senn hd449 on imac. I didn’t check how long does it takes but maybe one hour from start to silver then one hour and half or a little more for the gold one. The timbre section was without doubt the most ‘difficult’. While certain sessions, like loudness to say one, were too simple in my opinion.
I think philips made a good work. And it could be better -more difficulandt funny- with little correction. For exemple, Now I don’t remember where, but two different sample were played in sequence and when it changed there was written it was (hope you understand, sorry for my english). I suppose it was more funny if it played the two samples without saying when changes occour. Addendum: We needn’t argue if we follow a rule.
I’m always looking for a rule. I can appreciate how anyone would disagree with my opinion on something, but. You usually don’t follow me around (or I follow you around) and correct every opinion we don’t agree with. So I get the idea that this is different, and the reason you were so quick to express disagreement this time is because you can’t let my opinion to this new user go unchallenged for even a short time.
So I’m trying to understand why that is. Is it because you’re convinced that there’s no such thing as cancellation?
Or that I’m interpreting cancellation incorrectly? Or that I just imagined what I heard? Or is it because this project has some special importance that I’m potentially interfering with?
The technical aspects of this regarding cancellation of waveforms was never resolved as far as I know.
Hypothesis: Through 10 minute sessions of daily ear training I will increase the speed and accuracy of my pitch detection and EQ application.
Why: I’m sick of guessing and sweeping with the parametric EQ. I want to nail it every time. I don’t want to fear microphone feedback anymore. Ringing out stage monitors is a waste of time. I want to quickly remove any feedback as it occurs.
Results: After 30 days I made a 13% increase in my EQ ability and tripled the speed and accuracy of my pitch detection.
When it comes to listening abilities, I have always had a growth mindset. I don’t think I’ll ever have golden ears, but I do believe that I can train my ability up to a useful level. This is very important for me and all live sound engineers because we need shortcuts to survive.
During my Live Mix Mastery pilot course last year, I talked to a lot of sound engineers about their biggest problems out in the field. I got a variety of different answers, but the common trend among all of them was the need for speed. Everyone I talked to was confident that they could overcome any obstacle thrown at them if only they had enough time. As a result, I put together all of the best time saving techniques I’ve learned over the years and taught them to 20 students over four weeks.
Every technique I taught has been field tested to deliver results except for two things: pitch memory for feedback detection and EQ training for faster mixing.
With Live Mix Mastery I had a great opportunity to test this with a group of professional audio engineers. Here are the steps we took:
- User Ear Doctor in SoundGym to test your hearing.
- Schedule 10 minutes of daily ear training in your calendar. Three minutes playing Audio Frequency Trainer and seven minutes cycling through games on SoundGym.
Logically, playing these games to improve our ear training to increase our speed in the field makes sense. But I had never really taken the time to practice with a system and measure my results.
You have listened to your favorite song so many times that you can start singing it right now with pitch accuracy. Unless you were born with perfect pitch (yes, this exists) then you memorized those pitches through repetition. This is how the kid at guitar camp with me was able to identify almost any pitch. Songs were his reference. He had learned to play so many of them that playing any note would trigger his memory of a song and then its location on his fretboard. For me at the age of 18, this was mind-blowing.
I had my first taste of this in college when I set my wake up alarm to the song How It Feels to be Something On by Sunny Day Real Estate. One day I was walking into a piano rehearsal room, humming that song, sat down, and realized that I was singing a perfect A. By accident, I had taught myself pitch memory.
One of the first things you learn in music school is the interval relationships between notes on a scale in western music. Once you’ve got the pitch of any note, you can find the pitch of any other note through the memorized interval or by simply following a chromatic scale. The good news for musicians is that there are only 12 notes. The bad news for sound engineers is that microphone feedback could potentially happen at any frequency. And I guarantee you that it will never happen at the exact frequency of one of the sliders of your graphic EQ.
The only thing that graphic EQs are really good for is ear training, which is exactly what we used them for in Live Mix Mastery. Why did we use 1/3 octave spaced frequencies instead of 1/12 octave, which would relate more to our musical experience up until now? Three reasons:
- I didn’t think of it at the time.
- Audio Frequency Trainer was the best game I could find.
- Audio engineers are more familiar with the whole numbers seen on a graphic EQ. It’s a lot easier to work with 1K, 1.25K, and 1.6K than it is to work with 987.77, 1046.5, and 1174.66.
Audio Frequency Trainer will allow you to set a minimum and maximum test frequency, which is why we visited the Ear Doctor first. There are four levels that increase in difficulty by adding more frequencies to identify. I quickly moved out of Beginner, spent about two week on Intermediate, but never graduated from Pro. That shit is hard!
A technique that I used here, which I found helpful, was to move quickly and get emotional. My intention was to send signals to my brain’s pleasure and pain centers that this was important stuff.
Unfortunately, what I didn’t get to do was try out some feedback detection in the field, yet. I will come back and update this article when I do.
Another important thing I learned is that pitch memory either improves or deteriorates. I stopped practicing after the course ended and while I haven’t slide all the way back down to Beginner, I also haven’t been able to maintain a perfect score on Intermediate.
For many people, EQ is a big mystery. It’s one of the most difficult skills to train because we are always under enormous pressure. Wouldn’t we all love to have 30 minutes to listen to a kick mic while searching for the perfect frequencies to boost or cut. Those of you who have tried this have either never done so again, or moved into lighting.
EQ training at home is another thing that always made sense, logically, but I had never sat down to prove. Although none of these games we used are the same as work in the field with all of the chaos of a live room, they do provide the next best solution in terms of variety and tracking. Any time I have a few minutes I can log into SoundGym and play a game. At the moment, unfortunately, the games are not available for mobile, which is why I schedule my practice sessions for times when I know I will be home.
The great thing about this experiment that we embarked on together is that we didn’t have to worry about how to EQ. We just played the game and watched our results improve. The most enjoyable discovery for me was connecting the sounds I have known for years to specific frequencies. Previously I may have know where I needed to hear a filter, but would have had to guess and sweep up to it. The game Peak Master helped me to finally connect those sounds to frequencies. Here’s what one of my students, Sergio, said about it:
I had a big improvement detecting bothering or missing frequencies by ear.
And here’s what Martin said:
I was able to improve my skills to identify and remove distracting elements in my mix in less time. I no longer think, “Hmm, the electric guitar sounds weird somehow.” Now I can identify that the problem is in the low mids and make a dip at 300Hz, for example.
So it looks like we hit our goal in terms of increasing speed.
My big takeaway from this whole experience is to stop wondering how to EQ and improve my hearing instincts instead through ear training. Everyone knows when they hear a problem. The skill is finding it fast.
Although my students did see improvements in the field in increased speed and accuracy of pitch detection and EQ application, I personally haven’t done enough work to give a firm Yes. That being said, I’m really happy to have discovered a method I can track instead of just hoping for golden ears.
Popular Posts
Audio ear training for recording musicians and sound engineers. Get audio ear training online, improve core listening skills like frequency.
I like the idea but why should I train? The ability to connect what is in your mind with the appropriate parameters you have to dial to get that sound is not an easy task. The steps involved should be: Sometimes people get lost in the translation step and start turning knobs without confidence. The more you work, the better you understand what those knobs really do, but it is a slow process.
People excel in this matter after many years, because they have learned experimenting with lots of different processes applied to lots of different sources. The purpose of this training is to open your ears to what each frequency sounds like and reduce the amount of time needed to acquire this knowledge. In 15 minutes you can guess or correct 100 random equalisations, so training every day for a few weeks is equivalent to accumulating the experience of many years. Where does this training method come from?
Is the Golden Ears Audio Training Program any good? I\'ve heard good reviews from people that used it but Im not too sure. I seen it on Ebay for $50 so I wanted to know do you think it\'s worth it?
Did you make it up? No, we didn’t! The Guess Method has been used for over 40 years. There were exercises on cassettes, CD’s or it was done manually by teachers. We know it works and it has proven its efficiency over the years with thousands of successful engineers. The new Correct Method comes from a suggestion (the mastering guru) made to us: I suggest presenting a piece of recorded music which needs \'help\' in some frequency range and see if the student can not only identify the problem, but also correct for it in a musically satisfying way.
So that is what we did: correct instead of guess. We tested it and we loved it. It turned out to be both funny and instructive. Both methods complement each other really well. Besides, we have designed the software to turn these methods into a limitless tool, way better than any cassette or CD, because you can feed it with any music or sound and configure the exercises to a level of diversity and detail which wasn’t imaginable before. We can’t place more emphasis on this way of training the ear: IT WORKS!
Philips: Golden Ears Training A few months ago when Mike and I were at the in Belgium together with Tyll and Jude, Philips showed us their internal “Golden Ears” testing tool/program. I remember us all being very enthusiast about it, recommending them to make it available to the big public. And guess what? “We know that we are not alone in this obsession with sound, which is why we are inviting you to take our Golden Ears challenge”. The Golden Ears website has been online for a few months now even though Philips hasn’t really been marketing it.
That shows when you look at the number of finished tests on the website but I’m sure these numbers will increase quickly. So what is it exactly?
The golden Ears training program was designed to develop the listening skills of their audio engineers. Their expertise guarantees the creation of products of superior sound quality that highlight every musical detail.
Graduates of the program can reliably detect subtle differences in sound based on 5 key attributes: Timbre: Differences in tone or the frequency of content of music Details: The lack of treble, distortion, or noise that can mask the fine details Spatial Impression: The depth and spaciousness of the sound field Bass: The quality of low frequencies Loudness: The perception of sound power, from quiet to loud This online training program contains elements of their internal program we tried at their facility. To be fair, I do think their internal program is harder to complete than this online training program, but that seems only normal. This online version has four big levels, with short challenges in each of the above mentioned categories: Basic Level – Bronze Ears – Silver Ears – Golden Ears The thing I really liked about the training is that you get to experience a lot of the terms we use to describe sound in our reviews.
In example, if you sometimes wonder what a small sound stage or treble cut off sounds like, this is the place to learn. How long you will take to work your way to the end of the Golden Ears status is hard to predict but with an above average experience in sound like most of our readers have and a good dac/amp/headphone setup, you will (hopefully) go through it fairly quick. Don’t be mistaken however, there are some hard challenges in the Golden part for sure, especially the Timbre section.
Of course you have to start with the Basic level before you can go to the next level but you can do Loudness first and then go to Timbre and mix things up. If you answer any question incorrectly, you will need to go back one step. For your reference, there are 7 chapters in Basic Level; 6 chapters in Bronze Ears; 7 chapters in Silver Ears; 6 chapters and a quiz in the ultimate level – Golden Ears. After you’ve created a log in, the site will remember where exactly you were in the training so you can come back at any given time and continue where you quit. After completing a level you can share that with your friends on the social media. Who wouldn’t want to show off his Golden Ears status with his (audiophile) friends? My non audio friends of course now think of me as an even bigger freak, but what do they know, right?
I sincerely had fun doing all the tests using 2 different setups depending on my location: Cypher Labs Theorem DAC + Beyerdynamic A20 + Sennheiser HD650 and Meier Audio Corda Classic + Daccord + Hifiman HE-500. Philips recommends using the best headphones/setup you have available to complete the program. You can go have a look at, or even better, participate in the training program right here: Let us know if you have made it to the end and what you considered the hardest test was. To me the Timbre section in Gold was very tough. I hope you have as much fun as I had.
Thank you Philips for making this available to the big crowd, for free! (A mobile version of the site is coming soon) adrotate banner=”48″. If you see what I said about the colorations in the headphone you’re doing the tests with, and you’ve noted what Innerfidelity and others have said – that even flagships vary by up to plus or minus 5 db – then how do you propose to evaluate colorations that are smaller than your own headphone’s coloration? I’m going to guess in advance that you’re confident you can do that, and maybe you know a trick that works reliably.
But outside of using a Harman K812 or Sennheiser 800, I doubt the very premise when the differences in the test examples are small enough. And BTW, if the test really takes 10 times longer than I spent (5 sections times 2 – beginning and advanced), who is going to have that much free time? Experts I’d guess. I have to say I flew through everything except for Timbre on gold. Of course good gear is important.
I mentioned the gear I used in the article. The test might not be perfect and yes headphones can have variations, but if you know your headphone, you know what to listen to and you shouldn’t have to much problems IF your ears are good enough. That may seem like a weird statement but I have friends who can’t even make out a wider sound stage, some people just can’t hear those things so I suppose it will be harder for those trainees. I don’t know how Philips is generating the colorations, but based on my recent experience with Audioforge and how it can improve soundstage and eliminate the ‘constricted’ sound in closed headphones, I’d say they can generate those colorations with freq. Response only. And if that’s true, then a very neutral headphone would be needed.
The only reason I can think of that people can use very colored headphones in the training is when they’re lucky enough that a coloration in the headphone doesn’t coincide with a coloration in the test sample. Or, using the example from wave coincidence and reinforcement, that the headphone would cancel one real test track example and color another example that’s not colored. You know how it is that people are born seeing upside down and the brain has to reverse the picture?
That’s a potential problem with this training – you’ll train your brain to hear critically with a very colored headphone, which isn’t a good idea I don’t think. You should start with frequency response. Not only do many pricy headphones have peaks and recesses that you have to somehow compensate for in listening or judging, but frequency response deviations have a major impact on soundstage and other things, especially from the upper mids through the treble. One problem I see in Philips’ tests is the female voice, where the colorations they demonstrate involve frequency response deviations that are less than most headphones, then to make matters worse, the fundamentals where those deviations occur are often 10 to 20 db below the average signal. There are very brief moments where the fundamentals rise in volume, but I don’t know if it’s enough, especially given the colorations in our headphones. Another problem is the absolute secrecy as L said – we won’t know how any experts or reference people did on the tests, except to take their word for it (which I’d love to hear, but can’t make book on).
Generally, I’m trying to get feedback on my question mainly as something funthink of it as a hypothetical question, such as, what would I do if I won a million bucks:). I understand that you may not use scores or preset criteria, but was hoping to read some feedback as if you were given a class assignment to list 5 attributes you find most important to describe the sound quality of a product. I have an idea of what I would use.
Mainly I’m asking this because as a novice reviewer, I make short video reviews. Written reviews can literally be pages upon pages to describe the smallest nuances of how something sounds, which gives the writer a lot of real estate to review something in deep detail.
But, If I were to do a quick ‘nugget’ review in which I provide approximately 5 key attributes to quickly describe the most common aspects of what something sounds like, what would they be? This is of course a more subjective question, hence why I’m seeking more input. Also, I understand that people who review speakers, will perhaps key in on different attributes when compared to folks who generally only review IEM’s or full sized cans. My hearing is perfect.
Ear Training For Audio Engineers
But when someone is trying to whisper in my right ear and someone else is shouting in my left ear, I usually don’t hear the person on the right – not very clearly anyway. The way you claim you can hear perfectly through all of the colorations of your headphone to clearly hear the colorations in the samples – I don’t buy it.
It’s unscientific and illogical. Some you can hear, yes, but not all. Think about that some more and then if you have a logical explanation I’m sure interested. It just means that when two things are playing at the same time, there are cases where one sound “masks” the other sound, unless the 2 sounds are very different from each other and not too far apart in volume. In regard to the Philips training, if the headphone has enough coloration (i.e. The “really crappy” headphone one user noted), some of those colorations would coincide with colorations in the test music, and then correctly identifying the colored music sample would be just a guess. It’s like using a broken calculator that generates occasional wrong numbers when you’re building a bridge and making measurements.
I’m not sure if I understand you correctly, since english is not my first language, but all I’m saying is that I completed the test on gold difficulty with above mentioned headphones. But I never said that it was easy to do so – 2nd part of timbre test on gold took me like 2 hours. And I really don’t know what kind of “logical explanation” are you expecting from me? I simply passed the test and that’s all, but I can post screenshot from it + foto of my pair of HD 600’s / etys hf5 if it helps you somehow. The problem is, this is a new challenge for me, but much more important than that, it’s a major initiative that’s going to play a big role in defining the preferred sound of flagship (or lower) headphones going forward.
It’s not just Philips – Harman is working on this, and so is that pesky guy Paul Barton. So my interest is in finding exactly how people are hearing through one coloration to hear another, especially when (in some cases) it’s a similar coloration and one is going to mask the other. I’m not trying to give anyone a hard time, but we really need to understand some of the basics here instead of just “Hey – I heard it, well, maybe I did”. There is a training aspect to this, but training with very imperfect tools seems at least a little dubious to me.
I don’t buy it. And you can’t say something is plainly wrong just because Innerfidelity “proved” it wrong or right. What you can do IF you want to is demonstrate how you can have a headphone coloration that’s essentially the same as the coloration in the test track and “listen through” one to hear the other. You can’t demonstrate that by saying “I told you so”. And I don’t consider Innerfidelity to be any more than a testing lab. I find their measurements reasonably accurate and useful, but I don’t value their opinions as to what they hear – they have to stand in that line like everyone else.
And it’s a long line. I said it was wrong because I tried it on myself – do I have Stax or other expensive headphones? I have only HD 600 and etys hf5 with odac+o2.
Did I completed the test? Don’t know how for you, but for me it’s clear evidence that one don’t need super expensive headphones to complete that test.
Why it is so? I don’t know and to be honest I don’t really care, but just my 2 cents – you don’t need to care about headphones own coloration, you only need to focus for difference between two provided samples since that hp coloration is present in both samples. Sure there are exception to this, like bass/treble extension. Btw, I just finished bronze with super crappy Samsung earphones, going to try silver. Take the example of a headphone with output at 4 and 6 khz that’s -3 db compared to 5 khz. Then take a ‘colored’ music track that’s had its sound increased by 3 db at 4 and 6 khz.
Now that track will cancel out with the colored headphone and the uncolored track will exhibit the coloration from the headphone. Lemon tree fool\'s garden meaning. So in that particular case, what does training do? Teach us to hear upside down because we have a colored headphone?
In the real world, those examples I mention here don’t generally occur with the mathematical exactness I stated, but the principle is the same – the colored headphone will make some of the colored music sound better than the uncolored music, and that fact is proven in a hundred places. Beats is a large example, but there are many others. Philips isn’t doing this as a public service – this is going to feed into their ongoing designs, and that’s what I want to know more about – i.e., what’s the real goal for Philips? Well, exactly. Let’s just mark such a headphones as a “Headphone1”, uncolored music sample as A, colored sample as B. Wth are you talking about?
Golden Ears Audio Ear Training
How this deny EQ at all? You should either read again what I wrote or just stop trolling. And I was trying to explain it to you in numerous posts, but no matter what, you are still repeating the same thing that one can’t hear subtle differences because of headphone coloration, but without providing any real evidence to support your theory.
Said, his friend completed that test on gold with koss porta pro, and I did the same with hd600/hf5/samsung mobile earphones. I think you are just afraid to admit, that your hearing is not that good as you thought. It is pointless when you don’t respond to the point. So far, nobody has addressed the specific question as to how to hear a coloration that has been effectively canceled by an opposite coloration in the headphone. I suppose there’s an assumption that it would ‘rarely’ happen, but that’s a potential problem. In the early days of personal computers, virus scans, and a few other things, some of the ‘smart’ programs would look for certain byte signatures in target data to initiate an action, and if the particular signature were statistically very unlikely to impossible to happen otherwise, the programs would execute their action merrily without a second thought.
The WW2 Germans were confident of their Enigma machine’s security based on their statistics – wrong again! So today the success of certain people using the Philips training with very colored headphones has created a new wave of confidence that the headphone doesn’t matter, but I know it does matter, and so should you. Just take a very simple example of a headphone with a 5 db peak at 5 khz, then the test sample has a dip at 5 khz. They will cancel each other out. People can imagine all kinds of things, and even learn workarounds for some of these test things, but they can’t invent science. Waves reinforce and cancel, which is why forcing stereo to mono loses actual sounds, and why headphone crossfeed has made very little progress. To the extent that this training works at all, it’s because the colorations don’t coincide exactly.
But it isn’t good training to learn to detect colorations with a very colored headphone. Philips didn’t do this as a public service. They want to sell headphones, and not the colored kind.
Music Ear Training Free
The actual question is, how does one “train” to hear colorations that are heard with colored headphones, if the colorations in the music and the headphone coincide (either reinforce or cancel out, per ordinary wave physics)? 100 people can post their satisfactory experiences here, but we aren’t learning anything from that until someone does a critical analysis which explains it.
I know that there are forums where people relate their experiences through 100 to 1000 pages of postings that few people will ever read, but I’ve been hoping that on this excellent forum someone would dig into this and provide some real information. I completed the golden ears with senn hd449 on imac. I didn’t check how long does it takes but maybe one hour from start to silver then one hour and half or a little more for the gold one. The timbre section was without doubt the most ‘difficult’. While certain sessions, like loudness to say one, were too simple in my opinion.
I think philips made a good work. And it could be better -more difficulandt funny- with little correction. For exemple, Now I don’t remember where, but two different sample were played in sequence and when it changed there was written it was (hope you understand, sorry for my english). I suppose it was more funny if it played the two samples without saying when changes occour. Addendum: We needn’t argue if we follow a rule.
I’m always looking for a rule. I can appreciate how anyone would disagree with my opinion on something, but. You usually don’t follow me around (or I follow you around) and correct every opinion we don’t agree with. So I get the idea that this is different, and the reason you were so quick to express disagreement this time is because you can’t let my opinion to this new user go unchallenged for even a short time.
So I’m trying to understand why that is. Is it because you’re convinced that there’s no such thing as cancellation?
Or that I’m interpreting cancellation incorrectly? Or that I just imagined what I heard? Or is it because this project has some special importance that I’m potentially interfering with?
The technical aspects of this regarding cancellation of waveforms was never resolved as far as I know.
Hypothesis: Through 10 minute sessions of daily ear training I will increase the speed and accuracy of my pitch detection and EQ application.
Why: I’m sick of guessing and sweeping with the parametric EQ. I want to nail it every time. I don’t want to fear microphone feedback anymore. Ringing out stage monitors is a waste of time. I want to quickly remove any feedback as it occurs.
Results: After 30 days I made a 13% increase in my EQ ability and tripled the speed and accuracy of my pitch detection.
When it comes to listening abilities, I have always had a growth mindset. I don’t think I’ll ever have golden ears, but I do believe that I can train my ability up to a useful level. This is very important for me and all live sound engineers because we need shortcuts to survive.
During my Live Mix Mastery pilot course last year, I talked to a lot of sound engineers about their biggest problems out in the field. I got a variety of different answers, but the common trend among all of them was the need for speed. Everyone I talked to was confident that they could overcome any obstacle thrown at them if only they had enough time. As a result, I put together all of the best time saving techniques I’ve learned over the years and taught them to 20 students over four weeks.
Every technique I taught has been field tested to deliver results except for two things: pitch memory for feedback detection and EQ training for faster mixing.
With Live Mix Mastery I had a great opportunity to test this with a group of professional audio engineers. Here are the steps we took:
- User Ear Doctor in SoundGym to test your hearing.
- Schedule 10 minutes of daily ear training in your calendar. Three minutes playing Audio Frequency Trainer and seven minutes cycling through games on SoundGym.
Logically, playing these games to improve our ear training to increase our speed in the field makes sense. But I had never really taken the time to practice with a system and measure my results.
You have listened to your favorite song so many times that you can start singing it right now with pitch accuracy. Unless you were born with perfect pitch (yes, this exists) then you memorized those pitches through repetition. This is how the kid at guitar camp with me was able to identify almost any pitch. Songs were his reference. He had learned to play so many of them that playing any note would trigger his memory of a song and then its location on his fretboard. For me at the age of 18, this was mind-blowing.
I had my first taste of this in college when I set my wake up alarm to the song How It Feels to be Something On by Sunny Day Real Estate. One day I was walking into a piano rehearsal room, humming that song, sat down, and realized that I was singing a perfect A. By accident, I had taught myself pitch memory.
One of the first things you learn in music school is the interval relationships between notes on a scale in western music. Once you’ve got the pitch of any note, you can find the pitch of any other note through the memorized interval or by simply following a chromatic scale. The good news for musicians is that there are only 12 notes. The bad news for sound engineers is that microphone feedback could potentially happen at any frequency. And I guarantee you that it will never happen at the exact frequency of one of the sliders of your graphic EQ.
The only thing that graphic EQs are really good for is ear training, which is exactly what we used them for in Live Mix Mastery. Why did we use 1/3 octave spaced frequencies instead of 1/12 octave, which would relate more to our musical experience up until now? Three reasons:
- I didn’t think of it at the time.
- Audio Frequency Trainer was the best game I could find.
- Audio engineers are more familiar with the whole numbers seen on a graphic EQ. It’s a lot easier to work with 1K, 1.25K, and 1.6K than it is to work with 987.77, 1046.5, and 1174.66.
Audio Frequency Trainer will allow you to set a minimum and maximum test frequency, which is why we visited the Ear Doctor first. There are four levels that increase in difficulty by adding more frequencies to identify. I quickly moved out of Beginner, spent about two week on Intermediate, but never graduated from Pro. That shit is hard!
A technique that I used here, which I found helpful, was to move quickly and get emotional. My intention was to send signals to my brain’s pleasure and pain centers that this was important stuff.
Unfortunately, what I didn’t get to do was try out some feedback detection in the field, yet. I will come back and update this article when I do.
Another important thing I learned is that pitch memory either improves or deteriorates. I stopped practicing after the course ended and while I haven’t slide all the way back down to Beginner, I also haven’t been able to maintain a perfect score on Intermediate.
For many people, EQ is a big mystery. It’s one of the most difficult skills to train because we are always under enormous pressure. Wouldn’t we all love to have 30 minutes to listen to a kick mic while searching for the perfect frequencies to boost or cut. Those of you who have tried this have either never done so again, or moved into lighting.
EQ training at home is another thing that always made sense, logically, but I had never sat down to prove. Although none of these games we used are the same as work in the field with all of the chaos of a live room, they do provide the next best solution in terms of variety and tracking. Any time I have a few minutes I can log into SoundGym and play a game. At the moment, unfortunately, the games are not available for mobile, which is why I schedule my practice sessions for times when I know I will be home.
The great thing about this experiment that we embarked on together is that we didn’t have to worry about how to EQ. We just played the game and watched our results improve. The most enjoyable discovery for me was connecting the sounds I have known for years to specific frequencies. Previously I may have know where I needed to hear a filter, but would have had to guess and sweep up to it. The game Peak Master helped me to finally connect those sounds to frequencies. Here’s what one of my students, Sergio, said about it:
I had a big improvement detecting bothering or missing frequencies by ear.
And here’s what Martin said:
I was able to improve my skills to identify and remove distracting elements in my mix in less time. I no longer think, “Hmm, the electric guitar sounds weird somehow.” Now I can identify that the problem is in the low mids and make a dip at 300Hz, for example.
So it looks like we hit our goal in terms of increasing speed.
My big takeaway from this whole experience is to stop wondering how to EQ and improve my hearing instincts instead through ear training. Everyone knows when they hear a problem. The skill is finding it fast.
Although my students did see improvements in the field in increased speed and accuracy of pitch detection and EQ application, I personally haven’t done enough work to give a firm Yes. That being said, I’m really happy to have discovered a method I can track instead of just hoping for golden ears.
Loved this post? Join Sound Design Live and try these:...'>Golden Ears Audio Eartraining Program(20.04.2020)Audio ear training for recording musicians and sound engineers. Get audio ear training online, improve core listening skills like frequency.
I like the idea but why should I train? The ability to connect what is in your mind with the appropriate parameters you have to dial to get that sound is not an easy task. The steps involved should be: Sometimes people get lost in the translation step and start turning knobs without confidence. The more you work, the better you understand what those knobs really do, but it is a slow process.
People excel in this matter after many years, because they have learned experimenting with lots of different processes applied to lots of different sources. The purpose of this training is to open your ears to what each frequency sounds like and reduce the amount of time needed to acquire this knowledge. In 15 minutes you can guess or correct 100 random equalisations, so training every day for a few weeks is equivalent to accumulating the experience of many years. Where does this training method come from?
Is the Golden Ears Audio Training Program any good? I\'ve heard good reviews from people that used it but Im not too sure. I seen it on Ebay for $50 so I wanted to know do you think it\'s worth it?
Did you make it up? No, we didn’t! The Guess Method has been used for over 40 years. There were exercises on cassettes, CD’s or it was done manually by teachers. We know it works and it has proven its efficiency over the years with thousands of successful engineers. The new Correct Method comes from a suggestion (the mastering guru) made to us: I suggest presenting a piece of recorded music which needs \'help\' in some frequency range and see if the student can not only identify the problem, but also correct for it in a musically satisfying way.
So that is what we did: correct instead of guess. We tested it and we loved it. It turned out to be both funny and instructive. Both methods complement each other really well. Besides, we have designed the software to turn these methods into a limitless tool, way better than any cassette or CD, because you can feed it with any music or sound and configure the exercises to a level of diversity and detail which wasn’t imaginable before. We can’t place more emphasis on this way of training the ear: IT WORKS!
Philips: Golden Ears Training A few months ago when Mike and I were at the in Belgium together with Tyll and Jude, Philips showed us their internal “Golden Ears” testing tool/program. I remember us all being very enthusiast about it, recommending them to make it available to the big public. And guess what? “We know that we are not alone in this obsession with sound, which is why we are inviting you to take our Golden Ears challenge”. The Golden Ears website has been online for a few months now even though Philips hasn’t really been marketing it.
That shows when you look at the number of finished tests on the website but I’m sure these numbers will increase quickly. So what is it exactly?
The golden Ears training program was designed to develop the listening skills of their audio engineers. Their expertise guarantees the creation of products of superior sound quality that highlight every musical detail.
Graduates of the program can reliably detect subtle differences in sound based on 5 key attributes: Timbre: Differences in tone or the frequency of content of music Details: The lack of treble, distortion, or noise that can mask the fine details Spatial Impression: The depth and spaciousness of the sound field Bass: The quality of low frequencies Loudness: The perception of sound power, from quiet to loud This online training program contains elements of their internal program we tried at their facility. To be fair, I do think their internal program is harder to complete than this online training program, but that seems only normal. This online version has four big levels, with short challenges in each of the above mentioned categories: Basic Level – Bronze Ears – Silver Ears – Golden Ears The thing I really liked about the training is that you get to experience a lot of the terms we use to describe sound in our reviews.
In example, if you sometimes wonder what a small sound stage or treble cut off sounds like, this is the place to learn. How long you will take to work your way to the end of the Golden Ears status is hard to predict but with an above average experience in sound like most of our readers have and a good dac/amp/headphone setup, you will (hopefully) go through it fairly quick. Don’t be mistaken however, there are some hard challenges in the Golden part for sure, especially the Timbre section.
Of course you have to start with the Basic level before you can go to the next level but you can do Loudness first and then go to Timbre and mix things up. If you answer any question incorrectly, you will need to go back one step. For your reference, there are 7 chapters in Basic Level; 6 chapters in Bronze Ears; 7 chapters in Silver Ears; 6 chapters and a quiz in the ultimate level – Golden Ears. After you’ve created a log in, the site will remember where exactly you were in the training so you can come back at any given time and continue where you quit. After completing a level you can share that with your friends on the social media. Who wouldn’t want to show off his Golden Ears status with his (audiophile) friends? My non audio friends of course now think of me as an even bigger freak, but what do they know, right?
I sincerely had fun doing all the tests using 2 different setups depending on my location: Cypher Labs Theorem DAC + Beyerdynamic A20 + Sennheiser HD650 and Meier Audio Corda Classic + Daccord + Hifiman HE-500. Philips recommends using the best headphones/setup you have available to complete the program. You can go have a look at, or even better, participate in the training program right here: Let us know if you have made it to the end and what you considered the hardest test was. To me the Timbre section in Gold was very tough. I hope you have as much fun as I had.
Thank you Philips for making this available to the big crowd, for free! (A mobile version of the site is coming soon) adrotate banner=”48″. If you see what I said about the colorations in the headphone you’re doing the tests with, and you’ve noted what Innerfidelity and others have said – that even flagships vary by up to plus or minus 5 db – then how do you propose to evaluate colorations that are smaller than your own headphone’s coloration? I’m going to guess in advance that you’re confident you can do that, and maybe you know a trick that works reliably.
But outside of using a Harman K812 or Sennheiser 800, I doubt the very premise when the differences in the test examples are small enough. And BTW, if the test really takes 10 times longer than I spent (5 sections times 2 – beginning and advanced), who is going to have that much free time? Experts I’d guess. I have to say I flew through everything except for Timbre on gold. Of course good gear is important.
I mentioned the gear I used in the article. The test might not be perfect and yes headphones can have variations, but if you know your headphone, you know what to listen to and you shouldn’t have to much problems IF your ears are good enough. That may seem like a weird statement but I have friends who can’t even make out a wider sound stage, some people just can’t hear those things so I suppose it will be harder for those trainees. I don’t know how Philips is generating the colorations, but based on my recent experience with Audioforge and how it can improve soundstage and eliminate the ‘constricted’ sound in closed headphones, I’d say they can generate those colorations with freq. Response only. And if that’s true, then a very neutral headphone would be needed.
The only reason I can think of that people can use very colored headphones in the training is when they’re lucky enough that a coloration in the headphone doesn’t coincide with a coloration in the test sample. Or, using the example from wave coincidence and reinforcement, that the headphone would cancel one real test track example and color another example that’s not colored. You know how it is that people are born seeing upside down and the brain has to reverse the picture?
That’s a potential problem with this training – you’ll train your brain to hear critically with a very colored headphone, which isn’t a good idea I don’t think. You should start with frequency response. Not only do many pricy headphones have peaks and recesses that you have to somehow compensate for in listening or judging, but frequency response deviations have a major impact on soundstage and other things, especially from the upper mids through the treble. One problem I see in Philips’ tests is the female voice, where the colorations they demonstrate involve frequency response deviations that are less than most headphones, then to make matters worse, the fundamentals where those deviations occur are often 10 to 20 db below the average signal. There are very brief moments where the fundamentals rise in volume, but I don’t know if it’s enough, especially given the colorations in our headphones. Another problem is the absolute secrecy as L said – we won’t know how any experts or reference people did on the tests, except to take their word for it (which I’d love to hear, but can’t make book on).
Generally, I’m trying to get feedback on my question mainly as something funthink of it as a hypothetical question, such as, what would I do if I won a million bucks:). I understand that you may not use scores or preset criteria, but was hoping to read some feedback as if you were given a class assignment to list 5 attributes you find most important to describe the sound quality of a product. I have an idea of what I would use.
Mainly I’m asking this because as a novice reviewer, I make short video reviews. Written reviews can literally be pages upon pages to describe the smallest nuances of how something sounds, which gives the writer a lot of real estate to review something in deep detail.
But, If I were to do a quick ‘nugget’ review in which I provide approximately 5 key attributes to quickly describe the most common aspects of what something sounds like, what would they be? This is of course a more subjective question, hence why I’m seeking more input. Also, I understand that people who review speakers, will perhaps key in on different attributes when compared to folks who generally only review IEM’s or full sized cans. My hearing is perfect.
Ear Training For Audio Engineers
But when someone is trying to whisper in my right ear and someone else is shouting in my left ear, I usually don’t hear the person on the right – not very clearly anyway. The way you claim you can hear perfectly through all of the colorations of your headphone to clearly hear the colorations in the samples – I don’t buy it.
It’s unscientific and illogical. Some you can hear, yes, but not all. Think about that some more and then if you have a logical explanation I’m sure interested. It just means that when two things are playing at the same time, there are cases where one sound “masks” the other sound, unless the 2 sounds are very different from each other and not too far apart in volume. In regard to the Philips training, if the headphone has enough coloration (i.e. The “really crappy” headphone one user noted), some of those colorations would coincide with colorations in the test music, and then correctly identifying the colored music sample would be just a guess. It’s like using a broken calculator that generates occasional wrong numbers when you’re building a bridge and making measurements.
I’m not sure if I understand you correctly, since english is not my first language, but all I’m saying is that I completed the test on gold difficulty with above mentioned headphones. But I never said that it was easy to do so – 2nd part of timbre test on gold took me like 2 hours. And I really don’t know what kind of “logical explanation” are you expecting from me? I simply passed the test and that’s all, but I can post screenshot from it + foto of my pair of HD 600’s / etys hf5 if it helps you somehow. The problem is, this is a new challenge for me, but much more important than that, it’s a major initiative that’s going to play a big role in defining the preferred sound of flagship (or lower) headphones going forward.
It’s not just Philips – Harman is working on this, and so is that pesky guy Paul Barton. So my interest is in finding exactly how people are hearing through one coloration to hear another, especially when (in some cases) it’s a similar coloration and one is going to mask the other. I’m not trying to give anyone a hard time, but we really need to understand some of the basics here instead of just “Hey – I heard it, well, maybe I did”. There is a training aspect to this, but training with very imperfect tools seems at least a little dubious to me.
I don’t buy it. And you can’t say something is plainly wrong just because Innerfidelity “proved” it wrong or right. What you can do IF you want to is demonstrate how you can have a headphone coloration that’s essentially the same as the coloration in the test track and “listen through” one to hear the other. You can’t demonstrate that by saying “I told you so”. And I don’t consider Innerfidelity to be any more than a testing lab. I find their measurements reasonably accurate and useful, but I don’t value their opinions as to what they hear – they have to stand in that line like everyone else.
And it’s a long line. I said it was wrong because I tried it on myself – do I have Stax or other expensive headphones? I have only HD 600 and etys hf5 with odac+o2.
Did I completed the test? Don’t know how for you, but for me it’s clear evidence that one don’t need super expensive headphones to complete that test.
Why it is so? I don’t know and to be honest I don’t really care, but just my 2 cents – you don’t need to care about headphones own coloration, you only need to focus for difference between two provided samples since that hp coloration is present in both samples. Sure there are exception to this, like bass/treble extension. Btw, I just finished bronze with super crappy Samsung earphones, going to try silver. Take the example of a headphone with output at 4 and 6 khz that’s -3 db compared to 5 khz. Then take a ‘colored’ music track that’s had its sound increased by 3 db at 4 and 6 khz.
Now that track will cancel out with the colored headphone and the uncolored track will exhibit the coloration from the headphone. Lemon tree fool\'s garden meaning. So in that particular case, what does training do? Teach us to hear upside down because we have a colored headphone?
In the real world, those examples I mention here don’t generally occur with the mathematical exactness I stated, but the principle is the same – the colored headphone will make some of the colored music sound better than the uncolored music, and that fact is proven in a hundred places. Beats is a large example, but there are many others. Philips isn’t doing this as a public service – this is going to feed into their ongoing designs, and that’s what I want to know more about – i.e., what’s the real goal for Philips? Well, exactly. Let’s just mark such a headphones as a “Headphone1”, uncolored music sample as A, colored sample as B. Wth are you talking about?
Golden Ears Audio Ear Training
How this deny EQ at all? You should either read again what I wrote or just stop trolling. And I was trying to explain it to you in numerous posts, but no matter what, you are still repeating the same thing that one can’t hear subtle differences because of headphone coloration, but without providing any real evidence to support your theory.
Said, his friend completed that test on gold with koss porta pro, and I did the same with hd600/hf5/samsung mobile earphones. I think you are just afraid to admit, that your hearing is not that good as you thought. It is pointless when you don’t respond to the point. So far, nobody has addressed the specific question as to how to hear a coloration that has been effectively canceled by an opposite coloration in the headphone. I suppose there’s an assumption that it would ‘rarely’ happen, but that’s a potential problem. In the early days of personal computers, virus scans, and a few other things, some of the ‘smart’ programs would look for certain byte signatures in target data to initiate an action, and if the particular signature were statistically very unlikely to impossible to happen otherwise, the programs would execute their action merrily without a second thought.
The WW2 Germans were confident of their Enigma machine’s security based on their statistics – wrong again! So today the success of certain people using the Philips training with very colored headphones has created a new wave of confidence that the headphone doesn’t matter, but I know it does matter, and so should you. Just take a very simple example of a headphone with a 5 db peak at 5 khz, then the test sample has a dip at 5 khz. They will cancel each other out. People can imagine all kinds of things, and even learn workarounds for some of these test things, but they can’t invent science. Waves reinforce and cancel, which is why forcing stereo to mono loses actual sounds, and why headphone crossfeed has made very little progress. To the extent that this training works at all, it’s because the colorations don’t coincide exactly.
But it isn’t good training to learn to detect colorations with a very colored headphone. Philips didn’t do this as a public service. They want to sell headphones, and not the colored kind.
Music Ear Training Free
The actual question is, how does one “train” to hear colorations that are heard with colored headphones, if the colorations in the music and the headphone coincide (either reinforce or cancel out, per ordinary wave physics)? 100 people can post their satisfactory experiences here, but we aren’t learning anything from that until someone does a critical analysis which explains it.
I know that there are forums where people relate their experiences through 100 to 1000 pages of postings that few people will ever read, but I’ve been hoping that on this excellent forum someone would dig into this and provide some real information. I completed the golden ears with senn hd449 on imac. I didn’t check how long does it takes but maybe one hour from start to silver then one hour and half or a little more for the gold one. The timbre section was without doubt the most ‘difficult’. While certain sessions, like loudness to say one, were too simple in my opinion.
I think philips made a good work. And it could be better -more difficulandt funny- with little correction. For exemple, Now I don’t remember where, but two different sample were played in sequence and when it changed there was written it was (hope you understand, sorry for my english). I suppose it was more funny if it played the two samples without saying when changes occour. Addendum: We needn’t argue if we follow a rule.
I’m always looking for a rule. I can appreciate how anyone would disagree with my opinion on something, but. You usually don’t follow me around (or I follow you around) and correct every opinion we don’t agree with. So I get the idea that this is different, and the reason you were so quick to express disagreement this time is because you can’t let my opinion to this new user go unchallenged for even a short time.
So I’m trying to understand why that is. Is it because you’re convinced that there’s no such thing as cancellation?
Or that I’m interpreting cancellation incorrectly? Or that I just imagined what I heard? Or is it because this project has some special importance that I’m potentially interfering with?
The technical aspects of this regarding cancellation of waveforms was never resolved as far as I know.
Hypothesis: Through 10 minute sessions of daily ear training I will increase the speed and accuracy of my pitch detection and EQ application.
Why: I’m sick of guessing and sweeping with the parametric EQ. I want to nail it every time. I don’t want to fear microphone feedback anymore. Ringing out stage monitors is a waste of time. I want to quickly remove any feedback as it occurs.
Results: After 30 days I made a 13% increase in my EQ ability and tripled the speed and accuracy of my pitch detection.
When it comes to listening abilities, I have always had a growth mindset. I don’t think I’ll ever have golden ears, but I do believe that I can train my ability up to a useful level. This is very important for me and all live sound engineers because we need shortcuts to survive.
During my Live Mix Mastery pilot course last year, I talked to a lot of sound engineers about their biggest problems out in the field. I got a variety of different answers, but the common trend among all of them was the need for speed. Everyone I talked to was confident that they could overcome any obstacle thrown at them if only they had enough time. As a result, I put together all of the best time saving techniques I’ve learned over the years and taught them to 20 students over four weeks.
Every technique I taught has been field tested to deliver results except for two things: pitch memory for feedback detection and EQ training for faster mixing.
With Live Mix Mastery I had a great opportunity to test this with a group of professional audio engineers. Here are the steps we took:
- User Ear Doctor in SoundGym to test your hearing.
- Schedule 10 minutes of daily ear training in your calendar. Three minutes playing Audio Frequency Trainer and seven minutes cycling through games on SoundGym.
Logically, playing these games to improve our ear training to increase our speed in the field makes sense. But I had never really taken the time to practice with a system and measure my results.
You have listened to your favorite song so many times that you can start singing it right now with pitch accuracy. Unless you were born with perfect pitch (yes, this exists) then you memorized those pitches through repetition. This is how the kid at guitar camp with me was able to identify almost any pitch. Songs were his reference. He had learned to play so many of them that playing any note would trigger his memory of a song and then its location on his fretboard. For me at the age of 18, this was mind-blowing.
I had my first taste of this in college when I set my wake up alarm to the song How It Feels to be Something On by Sunny Day Real Estate. One day I was walking into a piano rehearsal room, humming that song, sat down, and realized that I was singing a perfect A. By accident, I had taught myself pitch memory.
One of the first things you learn in music school is the interval relationships between notes on a scale in western music. Once you’ve got the pitch of any note, you can find the pitch of any other note through the memorized interval or by simply following a chromatic scale. The good news for musicians is that there are only 12 notes. The bad news for sound engineers is that microphone feedback could potentially happen at any frequency. And I guarantee you that it will never happen at the exact frequency of one of the sliders of your graphic EQ.
The only thing that graphic EQs are really good for is ear training, which is exactly what we used them for in Live Mix Mastery. Why did we use 1/3 octave spaced frequencies instead of 1/12 octave, which would relate more to our musical experience up until now? Three reasons:
- I didn’t think of it at the time.
- Audio Frequency Trainer was the best game I could find.
- Audio engineers are more familiar with the whole numbers seen on a graphic EQ. It’s a lot easier to work with 1K, 1.25K, and 1.6K than it is to work with 987.77, 1046.5, and 1174.66.
Audio Frequency Trainer will allow you to set a minimum and maximum test frequency, which is why we visited the Ear Doctor first. There are four levels that increase in difficulty by adding more frequencies to identify. I quickly moved out of Beginner, spent about two week on Intermediate, but never graduated from Pro. That shit is hard!
A technique that I used here, which I found helpful, was to move quickly and get emotional. My intention was to send signals to my brain’s pleasure and pain centers that this was important stuff.
Unfortunately, what I didn’t get to do was try out some feedback detection in the field, yet. I will come back and update this article when I do.
Another important thing I learned is that pitch memory either improves or deteriorates. I stopped practicing after the course ended and while I haven’t slide all the way back down to Beginner, I also haven’t been able to maintain a perfect score on Intermediate.
For many people, EQ is a big mystery. It’s one of the most difficult skills to train because we are always under enormous pressure. Wouldn’t we all love to have 30 minutes to listen to a kick mic while searching for the perfect frequencies to boost or cut. Those of you who have tried this have either never done so again, or moved into lighting.
EQ training at home is another thing that always made sense, logically, but I had never sat down to prove. Although none of these games we used are the same as work in the field with all of the chaos of a live room, they do provide the next best solution in terms of variety and tracking. Any time I have a few minutes I can log into SoundGym and play a game. At the moment, unfortunately, the games are not available for mobile, which is why I schedule my practice sessions for times when I know I will be home.
The great thing about this experiment that we embarked on together is that we didn’t have to worry about how to EQ. We just played the game and watched our results improve. The most enjoyable discovery for me was connecting the sounds I have known for years to specific frequencies. Previously I may have know where I needed to hear a filter, but would have had to guess and sweep up to it. The game Peak Master helped me to finally connect those sounds to frequencies. Here’s what one of my students, Sergio, said about it:
I had a big improvement detecting bothering or missing frequencies by ear.
And here’s what Martin said:
I was able to improve my skills to identify and remove distracting elements in my mix in less time. I no longer think, “Hmm, the electric guitar sounds weird somehow.” Now I can identify that the problem is in the low mids and make a dip at 300Hz, for example.
So it looks like we hit our goal in terms of increasing speed.
My big takeaway from this whole experience is to stop wondering how to EQ and improve my hearing instincts instead through ear training. Everyone knows when they hear a problem. The skill is finding it fast.
Although my students did see improvements in the field in increased speed and accuracy of pitch detection and EQ application, I personally haven’t done enough work to give a firm Yes. That being said, I’m really happy to have discovered a method I can track instead of just hoping for golden ears.
Loved this post? Join Sound Design Live and try these:...'>Golden Ears Audio Eartraining Program(20.04.2020)