> For the complete documentation index, see [llms.txt](https://academy.shade.inc/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://academy.shade.inc/assets/search-and-discovery-transcription-and-precision-filtering.md).

# Search and Discovery (Transcription & Precision Filtering)

## Transcription

Whenever you upload a video or audio file containing dialogue, Shade automatically generates a full transcription. Transcriptions are time-coded, searchable, and exportable, making it easy to navigate long recordings and find specific moments without scrubbing through footage manually.

### How It Works

After a video or audio file is uploaded, Shade detects whether it contains dialogue and, if so, queues a transcription job. The transcript is broken into segments tied to specific timestamps, and different voices are automatically identified and labeled as **Speaker A**, **Speaker B**, **Speaker C**, and so on.

Because transcription runs as a background job after upload, it may take a few minutes before the transcript is available on a newly uploaded file.

### Accessing the Transcript

Open any asset with dialogue and click the **Transcription** tab in the right panel. The full transcript is displayed as a list of labeled, time-coded segments. Clicking any segment jumps the video or audio playback directly to that point, making it easy to navigate by what was said rather than by timeline position.

<figure><img src="https://450378260-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FUNq6h2Bfw7aAK0J2ffnZ%2Fuploads%2FkorGP1sYBaLD39dnoTMP%2FScreenshot%202026-04-27%20at%205.41.59%E2%80%AFPM%20(1).png?alt=media&#x26;token=73b74996-d2f7-451c-9a7d-43877eb235ee" alt=""><figcaption></figcaption></figure>

### Searching Within a Transcript

A search bar at the top of the Transcription panel lets you search for specific words or phrases within that file's transcript. Matching segments are highlighted, so you can quickly scan through a long recording for the moments most relevant to you.

You can also download the transcript from the **Transcript options** (**…**) menu at the top of the panel.

### Exporting a Transcript

Transcripts can be exported in several formats to fit different workflows:

* **TXT** — Plain text, useful for reading, copying, or archiving dialogue.
* **SRT** — Subtitle format, compatible with most video players and editing tools.
* **VTT** — Web-native subtitle format, commonly used for web video captions.
* **ScriptSync** — For use with professional editorial workflows that support ScriptSync-formatted transcripts.

### Editing a Transcript

If you have edit access to a file, you can correct its transcript from the Transcription panel:

* **Edit a section:** Right-click a section and select **Edit section**. Make your changes, then press `Enter` to save or `Esc` to cancel.
* **Replace a word:** Right-click a word and select **Replace**. Enter the new word, then click **Replace** to change only that instance, or **Replace all** to change every instance in this file's transcript.
* **Add to dictionary:** Right-click a word and select **Add to dictionary** to help Shade get that term right in future transcriptions. You can manage your workspace's terms from **Dictionary** in the **Transcript options** menu. This requires edit access to the workspace.

Edited sections are marked **(Edited)**. Click the eye icon (**See Original**) next to the label to compare with the original text, then choose **Keep edited** or **Revert to Original**. To view the whole original transcript, select **See original transcript** from the **Transcript options** menu.

### Searching Transcriptions Globally

Transcription content is also indexed in Shade's global AI search. To search across all transcribed files in your drive for a specific word or phrase, wrap your query in quotes.

For example, searching **"basketball"** will return every video and audio file where the word "basketball" is spoken.

### Things to Keep in Mind

* Transcription is only generated for files where dialogue is detected. Music, ambient audio, or silent video will not produce a transcript.
* Transcription runs after upload and may take a few minutes to appear. If the Transcription tab is empty on a newly uploaded file, check back shortly.
* Speaker labels (Speaker A, B, C, etc.) are assigned automatically based on voice detection. They are not linked to named individuals from the People feature.

## Find Similar Assets

To find content like an asset you already have, right-click the asset and select **Find Similar Assets**. Shade opens a **Similar to** page for that asset, showing similar assets from the same drive.

## Precision Filtering

Shade allows you to filter files by various intrinsic, inherited, or custom metadata attributes, including but not limited to:

* Names
* Creation dates
* Extensions
* Tags
* Status labels
* EXIF data
* IPTC data

These filters can be accessed from the top of any asset page as shown below. Filters can be combined to find the the assets that match all the attributes you have selected.


---

# 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://academy.shade.inc/assets/search-and-discovery-transcription-and-precision-filtering.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.
