# Statistical Parametric Speech Synthesis Incorporating Generative Adversarial Networks

arXiv: https://arxiv.org/abs/1709.08041 · HTML: https://mlsanity.com/paper/1709.08041/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Sep 2017.

## Built on by (most cited first, 16 of 16)

- [A Versatile Diffusion Transformer with Mixture of Noise Levels for Audiovisual Generation](https://mlsanity.com/paper/2405.13762/) (2024)
- [Audio representations for deep learning in sound synthesis: A review](https://mlsanity.com/paper/2201.02490/) (2022)
- [A Multi-Stage Multi-Codebook VQ-VAE Approach to High-Performance Neural TTS](https://mlsanity.com/paper/2209.10887/) (2022)
- [PeriodNet: A non-autoregressive waveform generation model with a structure separating periodic and aperiodic components](https://mlsanity.com/paper/2102.07786/) (2021)
- [Review of end-to-end speech synthesis technology based on deep learning](https://mlsanity.com/paper/2104.09995/) (2021)
- [How Deep Are the Fakes? Focusing on Audio Deepfake: A Survey](https://mlsanity.com/paper/2111.14203/) (2021)
- [DarkGAN: Exploiting Knowledge Distillation for Comprehensible Audio Synthesis with GANs](https://mlsanity.com/paper/2108.01216/) (2021)
- [Vocoder-Projected Feature Discriminator](https://mlsanity.com/paper/2508.17874/) (2025)
- [Provably Valid and Diverse Mutations of Real-World Media Data for DNN Testing](https://mlsanity.com/paper/2112.01956/) (2021)
- [Automated Side Channel Analysis of Media Software with Manifold Learning](https://mlsanity.com/paper/2112.04947/) (2021)
- [Generative models and Bayesian inversion using Laplace approximation](https://mlsanity.com/paper/2203.07755/) (2022)
- [Multi-Task Adversarial Training Algorithm for Multi-Speaker Neural Text-to-Speech](https://mlsanity.com/paper/2209.12549/) (2022)
- [Llama-VITS: Enhancing TTS Synthesis with Semantic Awareness](https://mlsanity.com/paper/2404.06714/) (2024)
- [HumanACGAN: conditional generative adversarial network with human-based auxiliary classifier and its evaluation in phoneme perception](https://mlsanity.com/paper/2102.04051/) (2021)
- [Improving novelty detection with generative adversarial networks on hand gesture data](https://mlsanity.com/paper/2304.06696/) (2023)
- [Generative Data Imputation for Sparse Learner Performance Data Using Generative Adversarial Imputation Networks](https://mlsanity.com/paper/2503.18982/) (2025)

## Abstract (arXiv)

A method for statistical parametric speech synthesis incorporating generative adversarial networks (GANs) is proposed. Although powerful deep neural networks (DNNs) techniques can be applied to artificially synthesize speech waveform, the synthetic speech quality is low compared with that of natural speech. One of the issues causing the quality degradation is an over-smoothing effect often observed in the generated speech parameters. A GAN introduced in this paper consists of two neural networks: a discriminator to distinguish natural and generated samples, and a generator to deceive the discriminator. In the proposed framework incorporating the GANs, the discriminator is trained to distinguish natural and generated speech parameters, while the acoustic models are trained to minimize the weighted sum of the conventional minimum generation loss and an adversarial loss for deceiving the discriminator. Since the objective of the GANs is to minimize the divergence (i.e., distribution difference) between the natural and generated speech parameters, the proposed method effectively alleviates the over-smoothing effect on the generated speech parameters. We evaluated the effectiveness for text-to-speech and voice conversion, and found that the proposed method can generate more natural spectral parameters and F0 than conventional minimum generation error training algorithm regardless its hyper-parameter settings. Furthermore, we investigated the effect of the divergence of various GANs, and found that a Wasserstein GAN minimizing the Earth-Mover's distance works the best in terms of improving synthetic speech quality.