DiffRhythm: Blazingly Fast and Embarrassingly Simple End-to-End Full-Length Song Generation with Latent Diffusion

Documenti analoghi

• DiffRhythm+: Controllable and Flexible Full-Length Song Generation with Preference Optimization
  di: Chen, Huakang, et al.
  Pubblicazione: (2025)
• DiffRhythm 2: Efficient and High Fidelity Song Generation via Block Flow Matching
  di: Jiang, Yuepeng, et al.
  Pubblicazione: (2025)
• SongEval: A Benchmark Dataset for Song Aesthetics Evaluation
  di: Yao, Jixun, et al.
  Pubblicazione: (2025)
• SynthVC: Leveraging Synthetic Data for End-to-End Low Latency Streaming Voice Conversion
  di: Guo, Zhao, et al.
  Pubblicazione: (2025)
• MeanVC: Lightweight and Streaming Zero-Shot Voice Conversion via Mean Flows
  di: Ma, Guobin, et al.
  Pubblicazione: (2025)