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

# Speaker

> Learn how speaker tags work in Vozo, when to adjust them, and how to improve voice results by managing speakers effectively.

## What Are Speakers?

Vozo automatically assigns speaker tags (e.g., Speaker 1, Speaker 2) based on voice characteristics. These tags help distinguish each person’s lines for translation, dubbing, and voice cloning.

<img src="https://mintcdn.com/vozo/H2aAjsZcH19aRL8G/images/speakertag.png?fit=max&auto=format&n=H2aAjsZcH19aRL8G&q=85&s=89013fb009a4a9010c1ff1c0298a8999" alt="Speakertag Pn" width="1762" height="182" data-path="images/speakertag.png" />

## How Speakers Are Detected

Vozo analyzes voice features like tone, pitch, and timing to detect different speakers. While detection is automatic, you can manually correct tags if needed.

## When to Manually Correct Speakers

Update or adjust speaker tags in these cases:

* A single speaker is mistakenly split into multiple tags.
* Multiple speakers are grouped under the same tag.
* A speaker’s voice varies dramatically in emotion (e.g., calm vs. angry). For lines with a distinct emotional tone, create a **new speaker** to help Vozo generate a more appropriate cloned voice.

## How to Add or Edit Speakers

<Steps>
  <Step title="Click the Speaker Tag">
    For the segment you want to fix, click the current speaker tag.
  </Step>

  <Step title="Select or Create Speaker">
    Choose the correct speaker from the dropdown, or click **New Speaker** to add a new one.
  </Step>

  <Step title="Update Dubbing">
    Once all corrections are made, click **Generate Speech** in the **top right corner** of the Speech section to apply the changes. A new cloned voice will be created for any newly added speakers.
  </Step>
</Steps>

## Simultaneous Speech

If multiple speakers say the same line at the same time, you can assign them to a single segment so their voices are dubbed together. See [Simultaneous Speech](/docs/translate_dub/simultaneous-speech) for details.

## Tips for Managing Speakers

* Rename speakers for clarity (e.g., "Host", "Narrator", "Guest").

<img src="https://mintcdn.com/vozo/TMbA1V5qn-lydk0P/images/image-7.png?fit=max&auto=format&n=TMbA1V5qn-lydk0P&q=85&s=f96df2ee403fce80d9b1146723adcb25" alt="Image" title="Image" style={{ width:"34%" }} width="440" height="490" data-path="images/image-7.png" />

* Use the **Filter** tool at the top-left of the Speech section to view and edit one speaker’s segments at a time.

<img src="https://mintcdn.com/vozo/H2aAjsZcH19aRL8G/images/speakerfilter.png?fit=max&auto=format&n=H2aAjsZcH19aRL8G&q=85&s=91e7c240c65bf306f685092e5cc82a92" alt="Speakerfilter Pn" width="1824" height="430" data-path="images/speakerfilter.png" />

* Use the **Clear Speaker** icon in the top-right corner of the **Change Speaker** dropdown to remove any speakers that are not assigned to any segment.
  <img src="https://mintcdn.com/vozo/TMbA1V5qn-lydk0P/images/image-8.png?fit=max&auto=format&n=TMbA1V5qn-lydk0P&q=85&s=18afd495e6551f9dd2f64fb163ced72b" alt="Image" title="Image" style={{ width:"33%" }} width="442" height="642" data-path="images/image-8.png" />

## FAQ

<AccordionGroup>
  <Accordion title="What if the cloned voice sounds wrong?">
    Double-check that the speaker tags are correct. If everything looks good but the voice still feels off, you can use the [Reclone Voice](/docs/translate_dub/reclone_voice) feature to generate a better result.
  </Accordion>

  <Accordion title="Can I reuse a speaker across projects?">
    Currently, **speaker tags are project-specific**, meaning each project detects and assigns speakers independently.\
    However, if you've cloned a voice for a speaker, you can **save it to your Library** and reuse it in other projects for consistent audio performance.
  </Accordion>
</AccordionGroup>