2 papers
cs.SD2025
IntMeanFlow: Few-step Speech Generation with Integral Velocity Distillation
Wei Wang, Rong Cao, Yi Guo +3
Flow-based generative models have greatly improved text-to-speech (TTS) synthesis quality, but inference speed remains limited by the iterative sampling process and multiple functi…
cs.LG2025
SplitMeanFlow: Interval Splitting Consistency in Few-Step Generative Modeling
Yi Guo, Wei Wang, Zhihang Yuan +8
Generative models like Flow Matching have achieved state-of-the-art performance but are often hindered by a computationally expensive iterative sampling process. To address this, r…