Unified Signal Compression Using a GAN with Iterative Latent Representation Optimization

Abstract

We propose a unified signal compression framework that uses a generative adversarial network (GAN) to compress heterogeneous signals. The compressed signal is represented as a latent vector and fed into a generator network that is trained to produce high quality realistic signals that minimize a target objective function. To efficiently quantize the compressed signal, non-uniformly quantized optimal latent vectors are identified by iterative back-propagation with alternating direction method of multipliers (ADMM) optimization performed for each iteration. The performance of the proposed signal compression method is assessed using multiple metrics including PSNR and MS-SSIM for image compression and also PESR, Kaldi, LSTM, and MLP performance for speech compression. Test results show that the proposed work outperforms recent state-of-the-art hand-crafted and deep learning-based signal compression methods.

arXiv comment13 pages, 10 figures

References in corpus (11)

• Neural Discrete Representation Learning
• A Systematic DNN Weight Pruning Framework using Alternating Direction Method of Multipliers
• Joint Autoregressive and Hierarchical Priors for Learned Image Compression
• Context-adaptive Entropy Model for End-to-end Optimized Image Compression
• Lossy Image Compression with Compressive Autoencoders
• GANSynth: Adversarial Neural Audio Synthesis
• Low Bit-Rate Speech Coding with VQ-VAE and a WaveNet Decoder
• Gradient Hard Thresholding Pursuit for Sparsity-Constrained Optimization
• Extremely Low Bit Neural Network: Squeeze the Last Bit Out with ADMM
• Training Skinny Deep Neural Networks with Iterative Hard Thresholding Methods
• Composition of Deep and Spiking Neural Networks for Very Low Bit Rate Speech Coding