> For the complete documentation index, see [llms.txt](https://docs.seegov.org/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.seegov.org/find-news/correct-speakers.md).

# Correct speaker names

Speaker correction is most useful when you plan to download the transcript or create a video. It helps the text you take from SeeGov—and the text SeeGov generates for your video—use a clearer name.

You do not need to identify every speaker just to review a meeting.

## Correct a speaker

1. Find the speaker label beside a transcript segment.
2. Choose its edit icon to open **Correct Speaker Information**.
3. Select the speaker type.
4. Search for an existing person or enter the requested name information.
5. Check the preview, then choose **Save**.

![The Correct Speaker Information dialog with speaker type, name, preview, and Unknown Speaker controls](https://1710228464-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FQCz5aNuL4AdSGk7ccQbq%2Fuploads%2Fgit-blob-6ee63d9f1e9465bf198b222233b4f52eae30d015%2Fcorrect-speaker.png?alt=media)

The corrected name appears anywhere that speaker ID is used in the meeting transcript.

Unnamed groups appear as **Unknown Speaker (ID 1)**, **Unknown Speaker (ID 2)**, and so on. These display IDs identify all speaker groups in the meeting, including named speakers; they do not count only unknown speakers and may differ from the raw transcription IDs. Named speakers also have a display ID, shown as **Speaker ID 3**, for example, below the correction dialog's title when you need to compare groups. Adding a name hides its ID in the transcript without renumbering other groups. Marking it unknown again restores the same label. Full and selected-moment downloads use the same display IDs as the meeting transcript.

## One speaker ID can cover many segments

Speech recognition assigns an ID to a detected voice, and that ID may appear in many transcript segments. Correcting it once can update every matching segment. That is helpful when the same council member speaks throughout a meeting.

One person can also have several IDs. Long recordings are transcribed in separate four-hour sections, each with its own IDs; numbering continues across the meeting, but correcting one group does not automatically correct that person's other groups. Numbers identify groups, not a verified count of people. Re-transcribing a meeting or changing its speaker groups can change the numbers.

Sometimes the system uses one ID for more than one person—for example, several public commenters at the same microphone. In that case, applying one person's name to every segment would make the transcript less accurate.

## When Unknown Speaker is safer

Choose **Mark as Unknown Speaker** when:

* the recording does not make the person's identity clear;
* more than one person appears to share the same speaker ID;
* the visible or spoken name conflicts with an earlier identification; or
* you are not confident enough to attach a real person's name.

An honest **Unknown Speaker** is better than a confident but incorrect attribution.

## What the correction changes

A correction improves the working transcript and can help downloaded transcripts, generated introductions, subtitles, and other text that refers to the speaker. It may also help SeeGov recognize the person in future meetings.

It does not change the source recording, and it does not prove the speaker's identity. Verify important attributions from the meeting, agenda, government roster, or another reliable source.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.seegov.org/find-news/correct-speakers.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
