Vyvo/Emilia-YODAS-ZH

Emilia-YODAS - ZH Clean version with only text and audio from the Emilia Dataset. Samples: 130,460 Language: ZH Usage from datasets import load_dataset dataset = load_dataset("Vyvo/Emilia-YODAS-ZH") sample = dataset['train'][0] text = sample['text'] audio = sample['audio']['array'] sampling_rate = sample['audio']['sampling_rate']