# CycleGAN-VC2: Improved CycleGAN-based Non-parallel Voice Conversion

arXiv: https://arxiv.org/abs/1904.04631 · HTML: https://mlsanity.com/paper/1904.04631/

## Key facts

- Cited by 46 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Apr 2019.

## Built on by (most cited first, 30 of 46)

- [Deepfakes Generation and Detection: State-of-the-art, open challenges, countermeasures, and way forward](https://mlsanity.com/paper/2103.00484/) (2021)
- [ContentVec: An Improved Self-Supervised Speech Representation by Disentangling Speakers](https://mlsanity.com/paper/2204.09224/) (2022)
- [An Overview of Affective Speech Synthesis and Conversion in the Deep Learning Era](https://mlsanity.com/paper/2210.03538/) (2022)
- [Dance Style Transfer with Cross-modal Transformer](https://mlsanity.com/paper/2208.09406/) (2022)
- [MaskCycleGAN-VC: Learning Non-parallel Voice Conversion with Filling in Frames](https://mlsanity.com/paper/2102.12841/) (2021)
- [Personalized Adversarial Data Augmentation for Dysarthric and Elderly Speech Recognition](https://mlsanity.com/paper/2205.06445/) (2022)
- [The Age of Synthetic Realities: Challenges and Opportunities](https://mlsanity.com/paper/2306.11503/) (2023)
- [Non-Autoregressive TTS with Explicit Duration Modelling for Low-Resource Highly Expressive Speech](https://mlsanity.com/paper/2106.12896/) (2021)
- [Style Equalization: Unsupervised Learning of Controllable Generative Sequence Models](https://mlsanity.com/paper/2110.02891/) (2021)
- [Learning the Beauty in Songs: Neural Singing Voice Beautifier](https://mlsanity.com/paper/2202.13277/) (2022)
- [Cross-Speaker Emotion Transfer for Low-Resource Text-to-Speech Using Non-Parallel Voice Conversion with Pitch-Shift Data Augmentation](https://mlsanity.com/paper/2204.10020/) (2022)
- [Data Redaction from Conditional Generative Models](https://mlsanity.com/paper/2305.11351/) (2023)
- [Many-to-Many Voice Conversion based Feature Disentanglement using Variational Autoencoder](https://mlsanity.com/paper/2107.06642/) (2021)
- [CycleGAN for Undamaged-to-Damaged Domain Translation for Structural Health Monitoring and Damage Detection](https://mlsanity.com/paper/2202.07831/) (2022)
- [Enhancing the Stability of LLM-based Speech Generation Systems through Self-Supervised Representations](https://mlsanity.com/paper/2402.03407/) (2024)
- [U Can't Gen This? A Survey of Intellectual Property Protection Methods for Data in Generative AI](https://mlsanity.com/paper/2406.15386/) (2024)
- [Towards General-Purpose Text-Instruction-Guided Voice Conversion](https://mlsanity.com/paper/2309.14324/) (2023)
- [Pathological voice adaptation with autoencoder-based voice conversion](https://mlsanity.com/paper/2106.08427/) (2021)
- [The Sound of Silence: Efficiency of First Digit Features in Synthetic Audio Detection](https://mlsanity.com/paper/2210.02746/) (2022)
- [Supervising 3D Talking Head Avatars with Analysis-by-Audio-Synthesis](https://mlsanity.com/paper/2504.13386/) (2025)
- [TGAVC: Improving Autoencoder Voice Conversion with Text-Guided and Adversarial Training](https://mlsanity.com/paper/2208.04035/) (2022)
- [iSTFTNet2: Faster and More Lightweight iSTFT-Based Neural Vocoder Using 1D-2D CNN](https://mlsanity.com/paper/2308.07117/) (2023)
- [VoxGenesis: Unsupervised Discovery of Latent Speaker Manifold for Speech Synthesis](https://mlsanity.com/paper/2403.00529/) (2024)
- [ReFlow-VC: Zero-shot Voice Conversion Based on Rectified Flow and Speaker Feature Optimization](https://mlsanity.com/paper/2506.01032/) (2025)
- [RT-VC: Real-Time Zero-Shot Voice Conversion with Speech Articulatory Coding](https://mlsanity.com/paper/2506.10289/) (2025)
- [An Adaptive Learning based Generative Adversarial Network for One-To-One Voice Conversion](https://mlsanity.com/paper/2104.12159/) (2021)
- [All-for-One and One-For-All: Deep learning-based feature fusion for Synthetic Speech Detection](https://mlsanity.com/paper/2307.15555/) (2023)
- [Reimagining Speech: A Scoping Review of Deep Learning-Powered Voice Conversion](https://mlsanity.com/paper/2311.08104/) (2023)
- [Axial Residual Networks for CycleGAN-based Voice Conversion](https://mlsanity.com/paper/2102.08075/) (2021)
- [A Novel Deep Learning Technique for Morphology Preserved Fetal ECG Extraction from Mother ECG using 1D-CycleGAN](https://mlsanity.com/paper/2310.03759/) (2023)

## Abstract (arXiv)

Non-parallel voice conversion (VC) is a technique for learning the mapping from source to target speech without relying on parallel data. This is an important task, but it has been challenging due to the disadvantages of the training conditions. Recently, CycleGAN-VC has provided a breakthrough and performed comparably to a parallel VC method without relying on any extra data, modules, or time alignment procedures. However, there is still a large gap between the real target and converted speech, and bridging this gap remains a challenge. To reduce this gap, we propose CycleGAN-VC2, which is an improved version of CycleGAN-VC incorporating three new techniques: an improved objective (two-step adversarial losses), improved generator (2-1-2D CNN), and improved discriminator (PatchGAN). We evaluated our method on a non-parallel VC task and analyzed the effect of each technique in detail. An objective evaluation showed that these techniques help bring the converted feature sequence closer to the target in terms of both global and local structures, which we assess by using Mel-cepstral distortion and modulation spectra distance, respectively. A subjective evaluation showed that CycleGAN-VC2 outperforms CycleGAN-VC in terms of naturalness and similarity for every speaker pair, including intra-gender and inter-gender pairs.