Amazon Lex is the AWS service designed to build conversational interfaces using voice and text. It provides the advanced deep learning functionalities of automatic speech recognition (ASR) for converting speech to text, and natural language understanding (NLU) to recognize the intent of the text, enabling developers to build applications with highly engaging user experiences and lifelike conversational interactions. It is the correct choice for the pharmaceutical company's requirements. Other options like SageMaker, Polly, and Transcribe have different primary functionalities: SageMaker is for model-building and machine learning, Polly is for text-to-speech capabilities, and Transcribe is for speech-to-text services.