metadata

datasets:
  - litagin/Galgame_Speech_ASR_16kHz
  - OOPPEENN/Galgame_Dataset
language:
  - ja
base_model:
  - kotoba-tech/kotoba-whisper-v2.0
library_name: transformers

~~Still Work In Progress and may be unstable!~~ I changed to use whisper-large-v3-turbo as a base model and currently training that, so this version will not be updated

Galgame Whisper (epoch 0.1)

Base: https://huggingface.co/kotoba-tech/kotoba-whisper-v2.0
Encoder frozen
Finetuned using: https://huggingface.co/datasets/litagin/Galgame_Speech_ASR_16kHz
Current epochs: 0.1 (13 / 114 tar files)
🤗 spaces Demo: https://huggingface.co/spaces/litagin/galgame_whisper_wip_demo
Maybe only works for Japanese