ayousanz/moe-speech-5speakers-ljspeech

moe-speech-5speakers-ljspeech Japanese multi-speaker TTS dataset in LJSpeech format. Overview Speakers: 5 Total utterances: 19,617 Format: LJSpeech (wavs/ + metadata.csv) Sample rate: 22050 Hz Language: Japanese Speaker Statistics Speaker ID Utterances 940de876 4,675 2cf01874 4,632 bbd90363 3,747 1a5a3db8 3,295 4e2f4ba6 3,268 Files metadata.csv - Full metadata (id|speaker|text) metadata_{speaker_id}.csv -… See the full description on the dataset page: https://huggingface.co/datasets/ayousanz/moe-speech-5speakers-ljspeech.