During the short external staff exchange from the second year of the project at the JSI partner, Octavian Machidon has worked closely with the JSI team with the goal of implementing a conversational intelligent agent capable of offering information about the European art using spoken natural language interaction. The results of this external staff exchange consist of a Google Assistant-based intelligent agent deployable on a Google Home or Android Smartphone device with the following features:
- Speech to text and text to speech support (in English)
- Visual feedback of the user’s query: the agent is displaying a list with the results showing pictures and description of the artwork, with a link to the original location
- The agent provides results from Europeana, a large collection of European art heritage
- The infrastructure for this intelligent agent is based on API.AI, Actions on Google, and a custom developed Webhook (a node.js web service) that acts as an intermediate layer between the agent and Europeana database
The agent is currently undergoing tests in the simulator, and once being approved by Google it will be published online in the following weeks.
