Intent Detection is a key component of any task-oriented conversational system. To understand the user's current goal and provide the most adequate response, the system must leverage its intent detector to classify the user's utterance into one of several predefined classes (intents). This objective can also simplify the set of processes that a conversational system must complete by performing the natural language understanding and dialog management tasks into a single process conducted by the intent detector. This is particularly useful for systems oriented to FAQ services. In this paper we present a novel approach for intent detection and classification based on word-embeddings and recurrent neural networks. We have validated our approach with a selection of the corpus acquired with the Hispabot-Covid19 system obtaining satisfactory results.