Skip to main content
Particle’ Podcast Intelligence indexes 100,000+ shows, fully transcribed, diarized, speaker-identified, labeled, and enriched with metadata within minutes of airing, making spoken conversations searchable. Each result is a speaker-attributed transcript window with timestamps.

​
Use it for

  • Finding expert commentary and quotable soundbites.
  • Media and brand monitoring.
  • Narrative and sentiment research.
  • Discovering and staying up to date with podcasts.

​
Provider ID

Use this value in dataSources: 
particle_news

​
Example

Find what podcast hosts are saying about AI regulation. 
from exa_py import Exa

exa = Exa()
run = exa.agent.runs.create(
    query="What are prominent podcast hosts and guests saying about AI regulation in 2025?",
    data_sources=[{"provider": "particle_news"}],
    output_schema={
        "type": "object",
        "required": ["mentions"],
        "properties": {
            "mentions": {
                "type": "array",
                "maxItems": 10,
                "items": {
                    "type": "object",
                    "required": ["podcast", "episode", "speaker", "quote", "stance"],
                    "properties": {
                        "podcast": {"type": "string"},
                        "episode": {"type": "string"},
                        "speaker": {"type": "string"},
                        "quote": {"type": "string"},
                        "stance": {"type": "string", "description": "pro-regulation, anti-regulation, or nuanced"},
                    },
                },
            }
        },
    },
)
run = exa.agent.runs.poll_until_finished(run.id)

​
Pairs well with

  • Financial Datasets: cross-check podcast chatter against published news.
  • Fiber.ai: attach company and contact context to the people being discussed.