# Comparing Representations for Audio Synthesis Using Generative Adversarial Networks

arXiv: https://arxiv.org/abs/2006.09266 · HTML: https://mlsanity.com/paper/2006.09266/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Jun 2020.

## Built on by (most cited first, 14 of 14)

- [Music2Latent: Consistency Autoencoders for Latent Audio Compression](https://mlsanity.com/paper/2408.06500/) (2024)
- [Musika! Fast Infinite Waveform Music Generation](https://mlsanity.com/paper/2208.08706/) (2022)
- [KAD: No More FAD! An Effective and Efficient Evaluation Metric for Audio Generation](https://mlsanity.com/paper/2502.15602/) (2025)
- [Evaluating generative audio systems and their metrics](https://mlsanity.com/paper/2209.00130/) (2022)
- [Survey on the Evaluation of Generative Models in Music](https://mlsanity.com/paper/2506.05104/) (2025)
- [CoDiCodec: Unifying Continuous and Discrete Compressed Representations of Audio](https://mlsanity.com/paper/2509.09836/) (2025)
- [DarkGAN: Exploiting Knowledge Distillation for Comprehensible Audio Synthesis with GANs](https://mlsanity.com/paper/2108.01216/) (2021)
- [Towards Controllable Audio Texture Morphing](https://mlsanity.com/paper/2304.11648/) (2023)
- [Music2Latent2: Audio Compression with Summary Embeddings and Autoregressive Decoding](https://mlsanity.com/paper/2501.17578/) (2025)
- [VQCPC-GAN: Variable-Length Adversarial Audio Synthesis Using Vector-Quantized Contrastive Predictive Coding](https://mlsanity.com/paper/2105.01531/) (2021)
- [Deep Neural Networks and End-to-End Learning for Audio Compression](https://mlsanity.com/paper/2105.11681/) (2021)
- [Data-Driven Modeling of Noise Time Series with Convolutional Generative Adversarial Networks](https://mlsanity.com/paper/2207.01110/) (2022)
- [SpecSinGAN: Sound Effect Variation Synthesis Using Single-Image GANs](https://mlsanity.com/paper/2110.07311/) (2021)
- [MARS: Sound Generation via Multi-Channel Autoregression on Spectrograms](https://mlsanity.com/paper/2509.26007/) (2025)

## Abstract (arXiv)

In this paper, we compare different audio signal representations, including the raw audio waveform and a variety of time-frequency representations, for the task of audio synthesis with Generative Adversarial Networks (GANs). We conduct the experiments on a subset of the NSynth dataset. The architecture follows the benchmark Progressive Growing Wasserstein GAN. We perform experiments both in a fully non-conditional manner as well as conditioning the network on the pitch information. We quantitatively evaluate the generated material utilizing standard metrics for assessing generative models, and compare training and sampling times. We show that complex-valued as well as the magnitude and Instantaneous Frequency of the Short-Time Fourier Transform achieve the best results, and yield fast generation and inversion times. The code for feature extraction, training and evaluating the model is available online.