Skip to content

Dataset · thivux

Large Vietnamese speech corpus for zero-shot TTS

thivux/phoaudiobook

Millions of Vietnamese audio-text pairs for training text-to-speech models without speaker fine-tuning.

What's inside

This is a large-scale dataset containing 1 to 10 million entries of Vietnamese speech and corresponding texts. The set is specifically prepared for zero-shot text-to-speech tasks, and its assembly methodology is described in a research paper for the ACL 2025 conference.

What you can do with it

  • Training zero-shot TTS models for the Vietnamese language
  • Training automatic speech recognition (ASR) systems in Vietnamese
  • Research in voice cloning
Tasks
Text → speech
Languages
Vietnamese
Size
1M–10M rows
Downloads this month
3.1K
Likes
57
Access
on request on Hugging Face

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("thivux/phoaudiobook")