Data

Overview

Following are the data objects available.

Document

A document represents a text-converted artifact.

It has the following attributes:

• id: ID of the document
• content: Text content. For an image, text content is obtained using OCR. For an audio file, text content is obtained using speech-to-text
• metadata: Additional data on this document, such as source (an url or a file path). For chunks, source can include the chunk index.

Embeddings

TBD

Vector Store

TBD