> ## Documentation Index
> Fetch the complete documentation index at: https://docs.runable.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Transcription

> Convert speech in audio or video files to text with speaker labels and audio event detection.

Transcription converts spoken content in audio or video files into written text. The AI identifies different speakers, labels them, and tags audio events like music, applause, or laughter.

## How to transcribe

<Steps>
  <Step title="Upload your file">
    Attach an audio or video file using the **+** button in the input bar. Supported formats include MP3, WAV, MP4, MOV, and other common audio and video formats.

    <Prompt description={`Transcribe a meeting recording.`} actions={["copy"]}>
      Transcribe this recording. Identify each speaker separately.
    </Prompt>
  </Step>

  <Step title="Wait for processing">
    The AI processes the file and generates a text transcript. Longer files take more time.
  </Step>

  <Step title="Review the transcript">
    The agent returns the full text with speaker labels. Each segment is tagged with the speaker who said it.

    <Frame caption="A transcription result showing speaker-labeled text in the chat.">
      ![Chat showing a transcription result with Speaker 1 and Speaker 2 labels on alternating paragraphs](https://pub-51091dcf1e9d4b04bb2e74f489c4f346.r2.dev/046ecc9ebb0b4126dab10cb09a79721d7149a198241960ba44f2872c24b17aee.png)
    </Frame>
  </Step>
</Steps>

## What transcription includes

| Feature                     | Description                                                                                          |
| --------------------------- | ---------------------------------------------------------------------------------------------------- |
| **Speaker diarization**     | The AI identifies different speakers and labels each segment (Speaker 1, Speaker 2, etc.).           |
| **Audio event tagging**     | Non-speech events like music, applause, laughter, and background noise are tagged in the transcript. |
| **Audio and video support** | Works with both audio files (MP3, WAV) and video files (MP4, MOV).                                   |

## When to use this

* Transcribing meeting recordings for written notes.
* Converting podcast episodes to text for blog posts or show notes.
* Creating subtitles or captions from video content.
* Extracting dialogue from video files for editing or analysis.

## What you cannot do

* You cannot transcribe in real time. Upload a complete file.
* You cannot assign custom names to detected speakers. Speakers are labeled numerically (Speaker 1, Speaker 2).
* You cannot transcribe content in multiple languages within the same file. The AI processes one language at a time.
* You cannot edit the transcript within Runable. Copy the text and edit it in your preferred text editor.

## Next steps

<CardGroup cols={2}>
  <Card title="AI Audio overview" icon="headphones" href="/overview-what-is-ai-audio">
    See all audio tools and capabilities.
  </Card>

  <Card title="Text-to-Speech" icon="microphone" href="/text-to-speech">
    Convert text into spoken audio.
  </Card>
</CardGroup>