Reference Workflows

Tokkio comes with several reference workflows for different use cases that can be used as a starting point. These reference workflows can then be further customized as described in the Customize Reference Workflows section as well as pages dedicated to a particular reference workflow.

Guidance on picking a workflow

The Tokkio LLM RAG workflow is simple, yet highly versatile and is highly recommended for most applications involving interactive virtual customer service assistants.

The LLM RAG workflow is demonstrated with three rendering options, which can be chosen according to the application needs:

1. Omniverse Renderer - Default option for Tokkio reference workflow. This rendering option enables you to set up metahuman avatars.

2. Live Portrait - Set up an avatar by simply providing a PNG portrait image file.

3. Unreal Engine Renderer - Realtime animation rendering from Unreal Engine Pixel Streaming through WebRTC

The Tokkio Retail workflow is suggested as an option for retail/e-commerce applications. Note that the rendering options described above can be applied for the retail workflow as well. An OV based rendering option for the retail workflow is provided as a reference.

• Tokkio LLM-RAG
  • Introduction
  • Source
    • Implementation details
  • RAG pipeline
    • Using a custom RAG pipeline
    • Adding filler sentences to bot response
    • Changing the avatar name and greeting
    • Updating bot gestures
    • Interruptions
    • Small Talk
    • Proactive bot
  • Limitations
    • Publish the bot
  • API Reference
  • Tokkio LLM RAG Rendering options
    • Tokkio LLM-RAG - Omniverse Renderer
      • Introduction
      • Minimum GPU Requirements
      • Source
      • Customization
    • Tokkio LLM-RAG - Unreal Engine
      • Introduction
      • Minimum GPU Requirements
      • Prerequisites
      • Architecture
      • Pre-steps
      • Deployment
    • Tokkio LLM-RAG - A2F-2D
      • Introduction
      • Minimum GPU Requirements
      • Prerequisite
      • Architecture
      • Deployment
      • Running the App
      • Customizations
      • Portrait picture customization
      • Animation Property Tuning
• Tokkio Retail
  • Introduction
  • Source
  • Minimum GPU Requirements
  • Deployment
  • Customizing the Retail Bot
  • Publishing the customized bot
    • Using customized Retail bot
  • Architecture
  • Sample Conversation flow with the Retail bot
    • ACE Agent Retail Plugin
    • Cart Manager
    • Catalog RAG
    • UI server and UI Front End
  • Tokkio Interaction Flow
    • Touch Based Interactions
    • Speech Based Interactions
    • Session Identification
  • Retail Workflow Limitations