Knowledge Base

VoxeNova's knowledge base lets you upload company documents so the AI facilitator has context during meetings. Documents are organised in a hierarchy and automatically indexed for retrieval.

Hierarchy

Content is organised in three levels:

Initiatives — top-level strategic themes (e.g. "Platform Modernisation")
Projects — concrete workstreams under an initiative (e.g. "API Gateway Redesign")
Documents — files attached to a project (PRDs, architecture docs, spreadsheets)

Creating Initiatives

      bash
      curl -X POST https://{your-slug}.voxenova.com/api/v1/initiatives \
  -H "Authorization: Bearer sk_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Platform Modernisation",
    "description": "Migrate monolith to microservices"
  }'
    

Creating Projects

      bash
      curl -X POST https://{your-slug}.voxenova.com/api/v1/projects \
  -H "Authorization: Bearer sk_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "API Gateway Redesign",
    "initiative_id": "{initiative_id}"
  }'
    

Uploading Documents

Upload files as multipart form data. Supported formats: PDF, DOCX, XLSX, images (PNG, JPG). Max size: 50 MB.

      bash
      curl -X POST https://{your-slug}.voxenova.com/api/v1/documents/upload \
  -H "Authorization: Bearer sk_your_api_key" \
  -F "file=@architecture-diagram.pdf" \
  -F "project_id={project_id}" \
  -F "title=Architecture Diagram v2"
    

Automatic processing. Uploaded documents are chunked, embedded, and indexed within seconds. Text is extracted from PDFs and images via OCR.

How Documents Feed Into Meetings

When you schedule a meeting linked to a project, VoxeNova uses the knowledge base in three ways:

Pre-meeting briefing — the AI reviews project documents to build context before the call starts
Entity extraction — names, technical terms, and acronyms from your docs are used to improve transcription accuracy
Real-time retrieval — during the meeting, relevant document passages are surfaced when topics align

Searching Documents

Run a semantic search across a project's knowledge base:

      bash
      curl -X POST https://{your-slug}.voxenova.com/api/v1/projects/{project_id}/search \
  -H "Authorization: Bearer sk_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "query": "authentication flow for mobile app",
    "top_k": 5
  }'
    

The response includes ranked document chunks with relevance scores and source references.

Next Steps

Schedule a meeting linked to your project
Choose the right meeting type for your use case