> For clean Markdown of any page, append `.md` to the page URL.
> For a complete documentation index, see https://docs.sarvam.ai/llms.txt.
> For full documentation content in one file, see https://docs.sarvam.ai/llms-full.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.sarvam.ai/_mcp/server.

# Browsing and Selecting Voices

Browse Sarvam's voice catalog and your own clones, preview any voice, and set it as your active persona. For an overview of the Voice Library and who it's for, see the [Voice Library overview](/creative-voice-library).

## Discovering Voices with Featured Voices

![Voice library screen. Header: Voice library title, Feedback button, and a black plus Clone voice button. Below it, a Featured voices section with category tabs (Audiobooks, Podcasts, Education, Marketing) above a row of stamp-style cards. Each card shows cover art, a title, and a Language, Narrator caption. Podcast-category cards pair two narrators per card.](https://files.buildwithfern.com/https://sarvam-api-docs.docs.buildwithfern.com/762a7e0625f5477984e22adb4f0e78d245e6597904c53665b45a22ad9e4cdb57/sarvam-creative/assets/voice-library-featured.png)

* **Featured voices** surfaces curated examples organized by content category - Audiobooks, Podcasts, Education, and Marketing.
* Switching category tabs swaps the example cards, each pairing a piece of sample content with the voice (or voices) narrating it.
* Podcast examples showcase two-narrator pairings - a useful preview of how two distinct personas sound side by side before you pick voices for a multi-speaker project.

## Browsing All Voices

![All voices section below Featured voices. Two tabs, My voices and Explore voices. Below the tabs, a search box and two filter dropdowns, both defaulted to All. The first filter lists use-case categories: Conversational, Audiobooks, Entertainment, Sales, News. The second filter lists gender: All, Male, Female. Below the filters, a table with columns Voices, Use case, and Gender. Each row shows a colored square play-button avatar, the voice's name, a short personality tag, a use-case chip, and a gender label. Hovering a row reveals a heart favorite icon and a Use voice button.](https://files.buildwithfern.com/https://sarvam-api-docs.docs.buildwithfern.com/31720ae1e8cbf7054493d4f7d5db8b8313491f1d2ff0f26cc80533c8fbede285/sarvam-creative/assets/voice-library-table.png)

Switch between the **My voices** and **Explore voices** tabs depending on whether you want your existing/favorited voices or the full catalog.

Narrow the table using the use-case filter (Conversational, Audiobooks, Entertainment, Sales, News) or the gender filter (All, Male, Female), or search by name.

* Worth the extra click once the library grows past a handful of voices - scanning by ear through dozens of unfiltered rows is slower than narrowing to the ones that actually match your project.

Click the play icon on any row to hear a short sample.

* Always preview before selecting for a real project - a voice's personality tag (e.g. "Warm & Reassuring") can suggest a tone the actual sample doesn't quite match.

Hover a row to reveal a heart icon (favorite it for quick access later) and a **Use voice** button (set it as your active persona).

* The selection carries over into Text to Speech and Dubbing automatically, so do this deliberately once you've settled on a voice rather than switching back and forth mid-project.

## Reading Voice Metadata

Each voice is tagged with a short personality descriptor, a use-case category, and a gender - for example:

| Voice | Personality       | Use Case       | Gender |
| ----- | ----------------- | -------------- | ------ |
| Shubh | Confident & Bold  | Conversational | Male   |
| Tanya | Friendly & Modern | Audiobooks     | Female |
| Anand | Warm & Reassuring | News           | Male   |
| Sunny | Cheerful & Upbeat | Sales          | Male   |
| Mani  | Calm & Composed   | Entertainment  | Male   |

Use the personality tag as a quick filter for tone, and the use-case tag as a starting point - a voice tagged "News" isn't locked to news content, it's just been curated with that context in mind.

## Sarvam Voices vs. Cloned Voices

* **Sarvam voices** - curated native-artist personas, ready to use immediately, no setup required.
* **Cloned voices** - voices you've created via Voice Cloning. These appear once processing finishes, and can be managed from the **My voices** tab.

See [Creating a Clone](/creative-cloning-create) to add a new voice to this library.