--- dataset_info: features: - name: audio dtype: audio - name: sentence dtype: string - name: length dtype: float64 splits: - name: train num_bytes: 6455680072.888 num_examples: 181408 download_size: 6403081821 dataset_size: 6455680072.888 configs: - config_name: default data_files: - split: train path: data/train-* license: artistic-2.0 task_categories: - automatic-speech-recognition - translation - text-to-speech - text-to-audio language: - ja tags: - Japanese - good dataset pretty_name: 'Audio and text from games. Japanese. Edited for NLP and ASR training. ' size_categories: - 100K