(Un)paired signal-to-signal translation with 1D conditional GANs
This is an incremental adaptation of existing image translation methods to 1D signals, potentially useful for audio or time-series processing.
The paper tackled the problem of unpaired signal-to-signal translation by adapting a 1D conditional GAN based on CycleGAN, and the result showed that noisy test signals could be transformed to resemble paired signals in the target domain, with differences quantified using correlation and error metrics.
I show that a one-dimensional (1D) conditional generative adversarial network (cGAN) with an adversarial training architecture is capable of unpaired signal-to-signal ("sig2sig") translation. Using a simplified CycleGAN model with 1D layers and wider convolutional kernels, mirroring WaveGAN to reframe two-dimensional (2D) image generation as 1D audio generation, I show that recasting the 2D image-to-image translation task to a 1D signal-to-signal translation task with deep convolutional GANs is possible without substantial modification to the conventional U-Net model and adversarial architecture developed as CycleGAN. With this I show for a small tunable dataset that noisy test signals unseen by the 1D CycleGAN model and without paired training transform from the source domain to signals similar to paired test signals in the translated domain, especially in terms of frequency, and I quantify these differences in terms of correlation and error.