Skip to content

Dataset · Moear

Azur Lane Character Voice Lines

Moear/AzurLane-Voice-20261118

Audio dataset of AzurLane voice lines in Japanese and Chinese for fine-tuning GPT-SoVITS.

What's inside

Includes Japanese and Chinese voice lines from the game AzurLane (碧蓝航线), organized via an SQLite database (about 163 MB) and TAR archives with audio files. The database provides character names, factions, ship types, original text, and localization. The dataset was used to train models like v2ProPlus.

What you can do with it

  • Training and fine-tuning GPT-SoVITS voice models
  • Text-to-speech (TTS) for specific game characters
Languages
Japanese, Chinese
Downloads this month
1K
Likes
4

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("Moear/AzurLane-Voice-20261118")