Author

Phannet Pov

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

#natural language process... Jun 2026

Closing the Quality Gap in Low-Resource Text-to-Speech: LoRA Fine-Tuning of VoxCPM2 for Khmer and Korean

VoxCPM2, a 2.4B parameter, tokenizer-free TTS model that joins a MiniCPM-4 language-model backbone with a flow-matching diffusion decoder, and the training loss and human ratings disagree on the best rank, shows that adaptation helps mainly where the base model is truly weak.

Phannet Pov, Hyung-Min Park, Voneat Pen et al. · arXiv.org · 1 citation