> For the complete documentation index, see [llms.txt](https://guide.scribe.gov.sg/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://guide.scribe.gov.sg/help/faqs/summaries.md).

# Summaries

{% hint style="danger" %}
**Scribe uses advanced audio/language processing techniques and will have inaccuracies in its outputs.**

Users are strongly advised to review the outputs generated by Scribe for accuracy before relying on it for critical tasks or decisions.
{% endhint %}

<details>

<summary><strong>How long are the summaries and transcripts stored for?</strong></summary>

**Summaries:** Summaries are kept for as long as their associated session. Once a session is deleted, its summaries are deleted too.

**Transcripts:** The generated transcripts are expired after 30 days by default. Users can configure this to other timeframes in user settings. [Learn how here.](/tips-and-tricks/configure-your-data-settings.md)

</details>

<details>

<summary><strong>Are the description and headers customisable for each session?</strong></summary>

Yes. While a default template is provided, you can customise the description and headers for each recording session and save them as custom templates. Learn more [here](https://guide.scribe.gov.sg/how-to.../save-templates).

The earlier templates used for your generated summaries will also appear under “Recently Used”, and you can make further tweaks as needed.

</details>

<details>

<summary><strong>Scribe sometimes misidentifies people or their roles in the summary. What can I do?</strong></summary>

This can sometimes happen because Scribe relies on context clues from the conversation to identify individuals and their roles. It may misinterpret these clues, especially if they’re not explicitly stated in the transcript.

To resolve this, you can generate a new summary and **provide more specific information in the description**. For example:

* *“I am an MSW speaking to the patient’s next-of-kin, Mrs Lim, about the patient’s discharge care plan.”*

This will **set better context for Scribe** and help Scribe better understand the conversation to generate more accurate summaries for you.

</details>

<details>

<summary><strong>When I use more headings, there seems to be less detail under each header that Scribe generates. Why?</strong></summary>

Scribe has a maximum number of words it'll generate based on the length of your transcript.

When you add more headers, Scribe might then generate less detail per section.&#x20;

To work around this, you could consider generating two different summaries with different headers. This way, you can cover all your desired headers while maintaining more detailed content under each.

</details>

<details>

<summary><strong>Is there a limit to the length of the descriptions and headers that I craft?</strong></summary>

No, but we'd encourage you to be concise when crafting your descriptions and headers as it would lead to better summaries.

</details>

<details>

<summary><strong>Can Scribe retain summaries in the original language, or are they all auto-generated in English?</strong></summary>

All summary outputs are automatically generated in English only.

</details>

<details>

<summary><strong>Can deleted summaries be recovered?</strong></summary>

No, deleted summaries cannot be recovered.

However, if the transcript is still available, you may be able to regenerate a similar summary by selecting the same template under **"Recently Used."**&#x20;

If the transcript is no longer available for your session, re-generating a new summary will not be possible.

</details>


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://guide.scribe.gov.sg/help/faqs/summaries.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.