particle_person_get_statement_characteristics

How a person speaks on podcasts, and how that moves over time. It reports the share of their lines on each axis:

Sentiment: negative, neutral or positive.
Certainty: hedged, neutral or confident.
Specificity: vague, neutral or specific.
Temporal orientation: about the past, the present or the future.

Each axis is shown for the whole window (the last year by default) and per week or month. Use it for questions like “is Sam Altman more cautious than he used to be?” or “how often does she make predictions?”. The labels are the ones speaker statement characteristics puts on transcript lines. Shares are of the person’s scored lines: lines spoken by them, in any role, that carry characteristics. Lines too short to read carry none. low_sample marks a share with fewer than 20 scored lines behind it. Read such a share as an example, not a distribution. To find their lines said a certain way about a topic, use particle_podcast_search_transcripts with person_slug and speaker_statement_characteristics. Resolve the person first with particle_person_resolve.

Inputs

Field	Type	Required	Default	Description
`person_slug`	string	yes	—	Canonical person slug from `particle_person_resolve` (e.g. `"sam-altman"`), or the encoded person ID.
`published_after`	string	no	a year before `published_before`	Start of the window (ISO 8601 date or date-time).
`published_before`	string	no	now	End of the window. A bare date includes that whole UTC day.
`interval`	enum	no	`month`	`week` (weeks start on Monday, UTC) or `month`.
`include`	array of enums	no	`[]`	`examples` adds each bucket’s strongest line for every outer label (negative/positive, hedged/confident, vague/specific, past/future), with its episode and a link that plays it.

Output

A markdown document headed ## How <name> speaks:

**Window:**, **Scored lines:** (with the number of episodes) and **Characteristics version:** rows.
### Overall: one bullet per axis, listing its labels in order with their shares.
### By month (or ### By week): a table with a row per bucket, zero-filled. Each row gives the scored lines and each axis’s label shares. A bucket the window cuts through is marked (partial). A bucket with fewer than 20 scored lines is marked (low sample).
With include: ["examples"], a ### Strongest lines, <bucket start> section per bucket. Each bullet reads **label** (axis): "text" — Episode on Podcast (episode_slug …) <link>.

The strongest line for a label is the one reading furthest past the label’s cut. Neither the readings nor the strength are exposed. Sample (person_slug="sam-altman", interval="month", include=["examples"], truncated):

## How Sam Altman speaks

**Window:** 2025-10-01 to 2026-10-27 (exclusive), by month
**Scored lines:** 9619 in 765 episodes
**Characteristics version:** 1

### Overall

- **Sentiment:** negative 10%, neutral 66%, positive 23%
- **Certainty:** hedged 27%, neutral 70%, confident 3%
- **Specificity:** vague 39%, neutral 57%, specific 4%
- **Temporal orientation:** past 17%, present 67%, future 16%

### By month

| Start | Lines | Sentiment | Certainty | Specificity | Temporal orientation |
| --- | --- | --- | --- | --- | --- |
| 2026-10-01 (partial) | 2528 | negative 12%, neutral 64%, positive 23% | hedged 28%, neutral 69%, confident 3% | vague 39%, neutral 57%, specific 4% | past 14%, present 72%, future 14% |

### Strongest lines, 2026-10-01

- **confident** (certainty): "We are clearly going to pick doing the right thing for humanity." — From Fair Game: Sam Altman Talks Elon Musk, ‘Artificial,’ and the Future of Humanity (Part 1 of 2) on Still Watching (episode_slug from-fair-game-sam-altman-talks-elon-musk-artificial-and-the-future-of-humanity) https://radar.particle.pro/podcasts/still-watching/episodes/from-fair-game-sam-altman-talks-elon-musk-artificial-and-the-future-of-humanity?t=1481.9

With output_format=json, the response is the REST body of GET /v1/people/{id}/statement-characteristics. It has:

scored_lines, episodes, low_sample and characteristics_version.
characteristics: one array per axis of {label, lines, share}.
buckets: each with the same fields, plus examples. Every example carries the line’s speaker_statement_characteristics and its episode.

Get started

Tools

particle_person_get_statement_characteristics

Inputs

Output

​Inputs

​Output

Inputs

Output