# An Improved StarGAN for Emotional Voice Conversion: Enhancing Voice Quality and Data Augmentation

arXiv: https://arxiv.org/abs/2107.08361 · HTML: https://mlsanity.com/paper/2107.08361/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 39 references that are in the map.
- First posted on arXiv 18 Jul 2021.

## Builds on (most cited first, 16 of 16)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Improved Training of Wasserstein GANs](https://mlsanity.com/paper/1704.00028/) (2017)
- [StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation](https://mlsanity.com/paper/1711.09020/) (2017)
- [AUTOVC: Zero-Shot Voice Style Transfer with Only Autoencoder Loss](https://mlsanity.com/paper/1905.05879/) (2019)
- [Seen and Unseen emotional style transfer for voice conversion with a new emotional speech dataset](https://mlsanity.com/paper/2010.14794/) (2020)
- [StarGAN-VC: Non-parallel many-to-many voice conversion with star generative adversarial networks](https://mlsanity.com/paper/1806.02169/) (2018)
- [Voice Conversion from Non-parallel Corpora Using Variational Auto-encoder](https://mlsanity.com/paper/1610.04019/) (2016)
- [F0-consistent many-to-many non-parallel voice conversion via conditional autoencoder](https://mlsanity.com/paper/2004.07370/) (2020)
- [Converting Anyone's Emotion: Towards Speaker-Independent Emotional Voice Conversion](https://mlsanity.com/paper/2005.07025/) (2020)
- [Attention-Augmented End-to-End Multi-Task Learning for Emotion Prediction from Speech](https://mlsanity.com/paper/1903.12424/) (2019)
- [Non-Parallel Voice Conversion with Cyclic Variational Autoencoder](https://mlsanity.com/paper/1907.10185/) (2019)
- [Unsupervised Representation Disentanglement using Cross Domain Features and Adversarial Learning in Variational Autoencoder based Voice Conversion](https://mlsanity.com/paper/2001.07849/) (2020)
- [Non-parallel Emotion Conversion using a Deep-Generative Hybrid Network and an Adversarial Pair Discriminator](https://mlsanity.com/paper/2007.12932/) (2020)
- [Augmenting Generative Adversarial Networks for Speech Emotion Recognition](https://mlsanity.com/paper/2005.08447/) (2020)
- [Investigation of F0 conditioning and Fully Convolutional Networks in Variational Autoencoder based Voice Conversion](https://mlsanity.com/paper/1905.00615/) (2019)
- [Voice Conversion Using Sequence-to-Sequence Learning of Context Posterior Probabilities](https://mlsanity.com/paper/1704.02360/) (2017)

## Built on by (most cited first, 7 of 7)

- [An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Era](https://mlsanity.com/paper/2210.03538/) (2022)
- [Nonparallel Emotional Voice Conversion For Unseen Speaker-Emotion Pairs Using Dual Domain Adversarial Network & Virtual Domain Pairing](https://mlsanity.com/paper/2302.10536/) (2023)
- [Reimagining Speech: A Scoping Review of Deep Learning-Powered Voice Conversion](https://mlsanity.com/paper/2311.08104/) (2023)
- [Disentangling segmental and prosodic factors to non-native speech comprehensibility](https://mlsanity.com/paper/2408.10997/) (2024)
- [Audio-based Kinship Verification Using Age Domain Conversion](https://mlsanity.com/paper/2410.11120/) (2024)
- [EmoAttack: Utilizing Emotional Voice Conversion for Speech Backdoor Attacks on Deep Speech Classification Models](https://mlsanity.com/paper/2408.15508/) (2024)
- [Neuron-Level Emotion Control in Speech-Generative Large Audio-Language Models](https://mlsanity.com/paper/2603.17231/) (2026)

## Abstract (arXiv)

Emotional Voice Conversion (EVC) aims to convert the emotional style of a source speech signal to a target style while preserving its content and speaker identity information. Previous emotional conversion studies do not disentangle emotional information from emotion-independent information that should be preserved, thus transforming it all in a monolithic manner and generating audio of low quality, with linguistic distortions. To address this distortion problem, we propose a novel StarGAN framework along with a two-stage training process that separates emotional features from those independent of emotion by using an autoencoder with two encoders as the generator of the Generative Adversarial Network (GAN). The proposed model achieves favourable results in both the objective evaluation and the subjective evaluation in terms of distortion, which reveals that the proposed model can effectively reduce distortion. Furthermore, in data augmentation experiments for end-to-end speech emotion recognition, the proposed StarGAN model achieves an increase of 2% in Micro-F1 and 5% in Macro-F1 compared to the baseline StarGAN model, which indicates that the proposed model is more valuable for data augmentation.