Spoken Dialogue System for Call Centers with Expressive Speech Synthesis

In this paper, we present a prototype of a spoken dialogue system that integrates automatic speech recognition (ASR), natural language understanding (NLU), bot management system, and expressive text-to-speech (TTS). Such a solution can be inte grated into a call center to provide first-line support, replace older interactive voice response (IVR) systems,decrease the load on call center operators and greatly improve client experience. The prototype is primarily designed for the Latvian lan guage, however, support for other languages can be easily added by replacing language-specific components like ASR and TTS.