So experiment with those parallelisms in octaves if you would. Now you could do the same thing if you’re going up to say E flat. You see you’re landing on the 7th chord or the A chord. For example my left hand’s playing C, B, B flat, A and my right hand a tenth above that E, E flat, D, C sharp. But if you go down a third or even if you go up a third for that matter you’ve got to do it chromatically. Another thing you can do is if you’re going down a third you can do it in triplets or take all three beats if you’re in 3/4 time. There’s a lots of rhythmic varieties to that obviously. Of course there’s many varieties to that. My left hand’s going down from C down to G. You can also do it : you probably wonder if I can go down. Triplet and then I’m there because I’m breaking the unit of time down into three segments. In other words if I went one, two, triple it, I could do it all on the third beat there couldn’t I? Or all on the fourth beat if I played it in triplets. If I was in 4/4 time then I’d have to start on B too wouldn’t I? One, two, three, four one : so it’s just a matter of timing because you obviously can’t get to where you’re going until it’s time to get there, so you want to delay that.Īnother thing you can do is do it in triplets. For example if I’m moving from C to F I can simply walk up in tenths or thirds. Now when can you use these parallel octaves? Whenever the chord moves up a fourth you can do it. My thumbs are only a third part and yet the rest of it, like it’s a tenth from my little finger to my right hand thumb and it’s an octave and a half from my little finger to my little finger. You could say it in thirds if you play your hands close enough together because my thumbs. If I play the C in octaves and the E in octaves that’s a parallel octaves in tenths. But if I take that E and move it up an octave that’s called a parallel tenth.
In other words if I play C and E together that’s a parallel third. What’s going on is that the left hand is playing an octave and the right hand is playing an octave and they’re going in the same direction. Those are parallel octaves that you’re hearing there.
Sometimes you’ve probably heard me in my teaching do things like this. Today I’d like to talk a little bit about parallel & contrary octaves in piano playing. This is Duane with some more good stuff you really ought to know. Here is a transcript of the podcast if you would like to follow along:ĭuane: Hello again.
Our code is available at: /chunmeifeng/Dual-OctConv.Parallel & Contrary Octaves in Piano Playing – How To Use Them The experimental results demonstrate the superiority of our model in accelerated parallel MR image reconstruction. Extensive experiments are conducted on an knee dataset under different undersampling patterns and acceleration factors. We evaluate the performance of the proposed model on the acceleration of multi-coil MR image reconstruction. Our framework provides two appealing benefits: (i) it encourages interactions between real and imaginary components at various spatial frequencies to achieve richer representational capacity, and (ii) it enlarges the receptive field by learning multiple spatial-frequency features of both the real and imaginary components. Then, our Dual-OctConv conducts intra-group information updating and inter-group information exchange to aggregate the contextual information across different groups. More specifically, the input feature maps and convolutional kernels are first split into two components (i.e., real and imaginary), which are then divided into four groups according to their spatial frequencies. By reformulating the complex operations using octave convolutions, our model shows a strong ability to capture richer representations of MR images, while at the same time greatly reducing the spatial redundancy. In this paper, we propose the Dual-Octave Convolution (Dual-OctConv), which is capable of learning multi-scale spatial-frequency features from both real and imaginary components, for fast parallel MR image reconstruction. Magnetic resonance (MR) image acquisition is an inherently prolonged process, whose acceleration by obtaining multiple undersampled images simultaneously through parallel imaging has always been the subject of research.