citedy-content-ingestion

You can convert YouTube videos and PDFs to structured text by submitting URLs for ingestion. The service processes each URL to return clean transcripts, metadata, and summaries suitable for immediate use in LLM pipelines.

Yes, you can batch process up to 20 URLs at once using the batch ingestion endpoint. This allows you to efficiently convert multiple web articles or media files into structured content objects with clear credit accounting.

Yes, you need a CITEDY_API_KEY set in your environment to transcribe audio files and extract web article content. The ingestion process uses specific REST endpoints to authenticate and manage credit accounting for each operation.

The best way to build a knowledge base from URLs is to use batch content ingestion. Submit up to 20 URLs containing PDFs, web articles, or audio files to extract clean text and metadata ready for indexing and summarization.

Yes, content ingestion supports extracting transcripts and metadata from audio files. By submitting audio file URLs for ingestion, you receive structured text outputs designed for downstream AI assistant workflows and content pipelines.

After URL ingestion, the structured content returns as an object containing clean text, transcripts, and metadata. This format is specifically designed to feed directly into AI assistants and downstream summarization pipelines.