Pigeon Voice Server Architecture - DatasmithSA/Pigeon-Voice-Training GitHub Wiki

The diagram below depicts the architecture of a typical Pigeon Voice Server implementation:

/images/Pigen-Voice-Architecture.png

  • Voice Console:
    • Vocollect Voice Console is used for managing devices, operators and Voice Tasks/Applications.
      • All communication between Vocollect Voice devices and Voice Console happen via port 9090.
      • Voice Console's database is hosted either on Microsoft SQL Server or Oracle. This database holds all the devices, operators and Tasks/Applications, as well as any other data used by Voice Console.
  • WMS: the WMS contains all business data relevant to all warehouse processes e.g. pick lists, warehouse locations, items etc.
  • Pigeon Voice Server: is the bridge between the Vocollect voice devices and the WMS in order to facilitate the transfer of the business data.
    • Request Handling: Vocollect Voice devices make requests to the Pigeon Voice Server; either via TCP/IP using LUT/ODR messages to port 15008/9 or via HTTP requests to port 80.
    • Custom Agents: through inversion of control the Pigeon Voice Server delegates a request to an appropriate custom agent (plugin) that you have developed. Each type of message being sent between the device and server requires a different agent to be developed.
      • Request: the agent will receive all information about the request from the device, such as:
        • Device ID
        • Operator ID
        • Device IP Address
        • Request Date/Time on Device
        • Input Records List
        • etc.
      • WMS Query: the agent will perform custom queries to gather the related business data out of the WMS. It is left up to the developer of the agents to determine how they query into the WMS e.g. SQL queries, calling of stored procedures etc. This allows for the flexibility to allow integration into any backend system.
      • Response: if the device requires a response, such as to a LUT request, the agent will populate the response with output records.
    • Response Handling: the Pigeon Voice Server will respond back to the Voice device with the response provided by the agent.
    • Logging: the Pigeon Voice Server will log all device requests and server responses to its own SQL Server database.