User-Adapted Semantic Description Generation Using Natural Language Models

Nowadays, conversational agents are making their way into our lives in many fields. The agent's speech is an important element in human-computer interaction, and to appear natural and friendly it should avoid predefined texts. With this premise, together with the significant growth in natural language generation models, this work explores the capabilities of natural language models to lead to a more fluent human-computer interaction and open up a range of new opportunities and applications. The present work proposes a system for the generation of descriptions tailored to the user's profile from different random topics using Natural Language models. After implementing the different applications, integrating natural language generation in conversational agents has proven to be highly effective.