Skip to main content
Most podcast APIs give you metadata — titles, descriptions, RSS feeds. Particle API gives you understanding. Every episode is transcribed, , structurally segmented, and linked to the knowledge graph. You get the transcript, every speaker identified by name and role, every topic classified, every entity extracted, and the most engaging moments surfaced as clips.

​
What you can build

  • Media monitoring — Track when a person, company, or topic is discussed across thousands of podcasts
  • Content discovery — Find the most engaging clips about any subject, ranked by AI engagement scoring
  • Research tools — Search across transcripts to find exactly what was said, by whom, and when
  • Podcast apps — Build players with chapters, speaker timelines, and highlight reels
  • Social content — Pull viral-ready clips with pre-written intro statements

​
Key concepts

Episodes

Full episodes with speakers, entities, topics, and enrichment counts.

Transcripts

Diarized dialogue, word-level timestamps, and SRT subtitle export.

Segments

AI-identified structural sections: topic discussions, interviews, ads, intros.

Clips

Engagement-scored highlight moments classified as funny, insightful, controversial, etc.

​
Resource hierarchy

Every resource is accessible both scoped to its parent (e.g., segments for a specific episode) and globally across all podcasts (e.g., all segments matching a filter). Scoped endpoints are standard tier; cross-podcast discovery endpoints are premium.

​
Pagination

All list endpoints use cursor-based pagination: 
{
  "data": [...],
  "has_more": true,
  "cursor": "r.4mGKaJ"
}
Pass cursor to the next request to get the next page. Cursors are opaque, compact tokens — treat them as strings and do not parse or construct them. The limit parameter controls page size (1–100, default 25).

​
Podcast speakers

The GET /v1/podcasts/{id} response includes a speakers array of recurring speakers across the podcast’s episodes. Each speaker has an aggregate occurrences count showing how many episodes they appear in.
Example (within a podcast response)
{
  "id": "2dMXSZK8XxM89hbS3eanr",
  "title": "Hard Fork",
  "speakers": [
    {
      "name": "Kevin Roose",
      "role": "HOST",
      "occurrences": 312,
      "entity_id": "Lm0wOWM3dzA"
    },
    {
      "name": "Casey Newton",
      "role": "CO_HOST",
      "occurrences": 305
    }
  ]
}
The entity_id links to the knowledge graph — use it with the entities API to find every podcast and episode where a speaker has appeared.