MASSIVE

The MASSIVE dataset is a comprehensive collection of approximately one million annotated utterances for various natural language understanding tasks such as slot-filling, intent detection, and Virtual Assistant performance evaluation.