Cycleflow: Leveraging Cycle Consistency In Flow Matching For Speaker Style Adaptation

Abstract

Voice Conversion (VC) aims to convert the style of a source speaker, such as timbre and pitch, to the style of any target speaker while preserving the linguistic content. However, the ground truth of the converted speech does not exist in a non-parallel VC scenario, which induces the train-inference mismatch problem. Moreover, existing methods still have an inaccurate pitch and low speaker adaptation quality, there is a significant disparity in pitch between the source and target speaker style domains. As a result, the models tend to generate speech with hoarseness, posing challenges in achieving high-quality voice conversion. In this study, we propose CycleFlow, a novel VC approach that leverages cycle consistency in conditional flow matching (CFM) for speaker timbre adaptation training on non-parallel data. Furthermore, we design a Dual-CFM based on VoiceCFM and PitchCFM to generate speech and improve speaker pitch adaptation quality. Experiments show that our method can significantl

Stats

• citations3
• S2 citations—
• github stars0
• HF likes0
• heat score4.52
• arxiv keyliang2025cycleflow

Related papers

• Enhancing Expressive Voice Conversion With Discrete Pitch-conditioned Flow Matching Model (2025)5.84
• Stablevc: Style Controllable Zero-shot Voice Conversion With Conditional Flow Matching (2024)7.81
• Zero-shot Voice Conversion Via Content-aware Timbre Ensemble And Conditional Flow Matching (2024)0.00
• Non-parallel Voice Conversion With Cyclic Variational Autoencoder (2019)12.10
• Delivering Speaking Style In Low-resource Voice Conversion With Multi-factor Constraints (2022)5.24
• Cross-lingual Text-to-speech With Flow-based Voice Conversion For Improved Pronunciation (2022)0.00
• Text-free Non-parallel Many-to-many Voice Conversion Using Normalising Flows (2022)7.16
• Parallel-data-free Voice Conversion Using Cycle-consistent Adversarial Networks (2017)0.00