VQCPC-GAN: VARIABLE-LENGTH ADVERSARIAL AUDIO SYNTHESIS USING VECTOR-QUANTIZED CONTRASTIVE PREDICTIVE CODING

Abstract

International audience ; Influenced by the field of Computer Vision, Generative Adversarial Networks (GANs) are often adopted for the audio domain using fixed-size two-dimensional spectrogram representations as the "image data". However, in the (musical) audio domain, it is often desired to generate output of variable duration. This paper presents VQCPC-GAN, an adversarial framework for synthesizing variablelength audio by exploiting Vector-Quantized Contrastive Predictive Coding (VQCPC). A sequence of VQCPC tokens extracted from real audio data serves as conditional input to a GAN architecture, providing step-wise time-dependent features of the generated content. The input noise z (characteristic in adversarial architectures) remains fixed over time, ensuring temporal consistency of global features. We evaluate the proposed model by comparing a diverse set of metrics against various strong baselines. Results show that, even though the baselines score best, VQCPC-GAN achieves comparable performance even when generating variable-length audio. Numerous sound examples are provided in the accompanying website, 1 and we release the code for reproducibility. 2 Index Terms-Generative Adversarial Networks, Audio Synthesis, Vector-Quantized Contrastive Predictive Coding * Nistal received funding from the European Union's Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No. 765068. 1 sonycslparis.github.io/vqcpc-gan.io 2 github.com/SonyCSLParis/vqcpc-gan