Skip to main content
Find dialogue lines where a specific person or company is named in podcast transcripts. Use this for the “every line about X” workflow. For dialogue that discusses a topic (paraphrase-tolerant or BM25), use particle_podcast_search_transcripts instead — that one ranks segments by relevance to a free-text query, not by who’s named.

​
Who said it

The guest_ filters keep only the mentions a matching guest spoke. guest_field="banking-finance-and-investment" with guest_standing="established" reads what established finance experts said about a company. When the question names who speaks (“doctors”, “people in finance”), resolve the words with particle_expertise_resolve first and filter by the code; on unresolved_reference, resolve and retry rather than dropping the filter. For the split of all mentions across occupations or fields, use particle_podcast_get_share_of_voice: each group it returns is a guest_occupation or guest_field value for this tool. All guest_ filters describe the same guest, so guest_occupation="cardiologists" with guest_practicing=true keeps lines spoken by a practicing cardiologist. A broad code covers everything below it. A word no occupation names may name a field or a capacity instead, and filters by it: guest_occupation="oncologists" is field 3211, guest_occupation="founders" is guest_capacity="executive_founder". An occupation or field that names nothing returns unresolved_reference with the closest codes as did-you-mean suggestions; guest_practicing without a detailed SOC 2018 guest_occupation returns validation_error.

​
Three response modes

format="summary" (default — wide scan). Returns up to limit episodes (reverse-chronological), each with metadata plus the first 10 mention-only lines (just the lines naming the entity, no surrounding dialogue). Use this to see what’s been said across episodes and decide which episodes are worth reading in full. Paginate older episodes with cursor. format="detail" (narrow drill-in). Requires episode_slug. Returns the full mention windows with context_lines of surrounding dialogue around each mention. Pass one slug for a single episode, or up to 10 comma-separated slugs (e.g. episode_slug="all-in-200,all-in-201,all-in-202") to multi-get several episodes in one call. limit/cursor don’t apply. format="compact" (screening). The same episodes as summary, each with its mention count, the strings it was mentioned as, and the segments carrying the mentions — id, title, type, mentions in that segment, and the time of the first one — with no dialogue lines at all. Attribution is per mention line, so a mention in a segment’s opening lines is credited to that segment even though its window (which opens context_lines earlier) began in the previous one. It is the smallest shape: use it to fan out over many entities or a long date range and decide where to read. The segment ids are citations; format="detail" with the episode slug reads the lines.

​
Workflow

  • No specific episode in mind: call format="summary" first to scan, then call format="detail" with the slug(s) of the episodes worth reading in full. For most questions (sentiment, recurring themes, who said what when), summary alone has enough signal and the second call isn’t needed.
  • Already have the episode slug (e.g. user mentioned the episode by name, or you got it from particle_podcast_get_episode / particle_podcast_search_transcripts): skip summary entirely and call format="detail" with episode_slug directly.

​
Inputs

One of person_slug or company_slug is required. Resolve a name to a slug first with particle_person_resolve, particle_entity_resolve, or particle_company_resolve.

​
Output

A markdown document with ## Mentions of <Name> (N episodes, format=<summary|detail|compact>) and one ### Episode title section per episode. Each section carries bulleted KV rows: - **Podcast:**, - **Podcast slug:**, - **Published:**, - **Episode slug:**, - **Mentions:**. In summary and detail mode, each episode first lists its identified speakers, one **Speaker:** Name (slug, ROLE) — Title (code) · standing · seniority row each. In summary mode, each episode is followed by a flat bullet list — one bullet per mention, formatted @ <seconds>s **Speaker:** text. When more pages exist, a horizontal rule and a **Cursor:** <value> line are appended; pass that value back as cursor to fetch the next page. In detail mode, each episode is followed by one bullet per mention window (Segment title (TYPE) @ start–end), then indented dialogue lines underneath. The line that actually names the entity has the entire Speaker: text wrapped in bold; surrounding context lines render as **Speaker:** text (speaker label bold, text plain). A line in a topic or interview segment ends with its speaker statement characteristics, pole labels only: _[confident · positive · future]_. _[neutral]_ marks a line that leans clearly on no axis, and a line without brackets carries no labels. With output_format=json, each line has the full speaker_statement_characteristics object. Sample (summary, person_slug="marc-andreessen", limit=2):
In compact mode, each episode is followed by a - **Mentioned as:** row and one bullet per segment carrying mentions, formatted Segment title (TYPE) — N mentions, first @ <seconds>s — segment id: <id>, in order of first mention. No dialogue is rendered. Pagination works as in summary mode. Sample (compact, company_slug="nvidia.com", since="2026-09-10", limit=2):
Sample (detail, drilling into one of those episodes):

​
Example

Who said it: what established finance experts said about NVIDIA.
Screening a watchlist: one compact call per company for the week, read only where the segment count and titles warrant it.