This dataset is sourced directly from ReazonSpeech. It corresponds to the tiny version of ReazonSpeech, containing 8.5 hours of audio data totaling approximately 600MB.
As the original dataset is incompatible with datasets (version >= 3.0), I downloaded and repackaged the dataset before re-uploading it to HuggingFace.
NOTE: The original data provider has stated that the specific use of this dataset is subject to Article 30-4 of the Japanese Copyright Act.
