Cantonese datasets

The catalogue lists 4 Cantonese datasets. Most listings are audio and text data, typically used for conversational AI, speech recognition and speech synthesis (TTS). Where providers state a volume in hours, it ranges from 57 hours to 2,162 hours.