A voice assistant for your site — answers from your materials

The scenario loads your documents from Box into the Postgres PGVector Store knowledge base and puts a voice assistant on your site: it hears the question, finds the answer and speaks it, staying within safe bounds.

  • Answers by voice from your documents
  • Runs on your site around the clock
  • Stays within safe bounds
  • Load materials once
Qdrant Vector StoreEmbeddings OpenAIWebhookGuardrailsDefault Data Loader
Connect Postgres PGVector Store and goScriptera

How it works

To connect Postgres PGVector Store and Box, you don't need a developer: a ready-made scenario links them in minutes.

  1. Starts when: Webhook
  2. Then: Generate text (STT)
  3. Then: Guardrails
  4. Then: Voicebot AI Agent
  5. Then: Generate audio (TTS)
  6. Starts when: Respond to Webhook
  7. Then: Default response (TTS)
  8. Starts when: Respond default to Webhook
  9. Then: OpenAI Chat Model
  10. Then: Window Buffer Memory
  11. Then: Calculator
  12. Then: Qdrant Vector Store
  13. Then: Embeddings OpenAI
  14. Then: OpenAI Chat Model1
  15. Then: OpenAI Chat Model2
  16. Then: RAG
  17. Starts when: When clicking ‘Test workflow’
  18. Then: Create collection
  19. Then: Clear collection
  20. Then: Search files
  21. Check: Loop Over Items
  22. Then: Get files
  23. Then: Insert file
  24. Then: Wait 5 sec.
  25. Then: Embeddings OpenAI1
  26. Then: Default Data Loader1
  27. Then: Recursive Character Text Splitter

You can launch this Postgres PGVector Store + Box integration in Scriptera: describe the task in plain words — the scenario is built, launched and monitored for you.

Similar scenarios