Skip to content

Dataset · XRailgunX

Voice lines of character Aemeath from Wuthering Waves

XRailgunX/Aemeath-Audio-Dataset

450 Chinese voice lines of a single character with transcripts and emotion labels for TTS training.

What's inside

Contains 450 Chinese audio recordings with a total duration of 49.21 minutes: 334 story lines and 116 combat/character phrases of the hero Aemeath from Wuthering Waves. Audio is decoded from Opus format to 48 kHz (422 mono and 28 stereo files). Source materials were taken from the Wuthering Waves Wiki, and the release authors corrected the transcripts and added emotion annotations.

What you can do with it

  • Training or fine-tuning (LoRA) a single-speaker TTS model in Chinese with emotion transfer
  • Testing speech recognition (ASR) and emotion classification models on game audio
Tasks
Text → speech, Automatic speech recognition, Audio classification
Languages
Chinese
Size
up to 1K rows
Downloads this month
3.1K
Likes
3

How to load it

from datasets import load_dataset

ds = load_dataset("XRailgunX/Aemeath-Audio-Dataset")