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

# Voice settings for text to speech

> Choose the voice that reads your script, switch between recent and favorite voices, set expressiveness from 0 to 2, and turn on up to 10 dictionaries.

The **Voice settings** tab beside the script decides how your script is read: which voice reads it, how expressive it sounds, and which dictionaries change how words are said. This page goes through each one, top to bottom.

![The Voice settings tab: the voice card for Rania, Change voice, the Recently used, Favorites and Project voices rows, three dictionaries on with Manage, and Expressiveness at 1](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/voice-settings-tab-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=dc3438a0b9548b42e8b638baedfbac0e)

## Choose a voice

The voice card at the top shows the voice that will read the script: its name, gender, dialect (or English), and its use cases. A clone made in your project has a "Custom voice" badge.

* Click the play button (**Play preview**) to hear the voice's sample.
* Click the star to add the voice to your favorites, or remove it. Clones have no star.
* Click **Change voice** to open the voice browser ("Select a voice"). Search and filter the voices, play samples, and click **Select** on the one you want. The [Voices library](https://docs.tryhamsa.com/media/voices/voices-library.md) explains the filters and tabs.

![The voice browser opened from Change voice, with tabs, search, language, dialect and use case filters, and voices each with a Select button](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/select-a-voice-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=cfa0fd457ed7faaea5d1ba9590d5ad4f)

Under the card, chips switch voices in one click:

| Row | What it holds |
| - | - |
| **Recently used** | The last 5 voices you generated with in this project, kept in this browser |
| **Favorites** | Your starred voices (the first 6). Only you see your favorites |
| **Project voices** | Voices cloned in this project (the first 6), for everyone in it |

The voice decides the language and the dialect of the speech. To get Egyptian Arabic, pick an Egyptian voice; English voices read English.

From the [Voices library](https://docs.tryhamsa.com/media/voices/voices-library.md) you can also click **Use** on a voice to open Text to speech with it.

### The voice on each visit

Each time you open Text to speech, the voice goes back to Hamsa's default voice, or to your project's default voice if you set one in [Project defaults](https://docs.tryhamsa.com/media/workspace/project-defaults.md). **More actions** (the three dots on the voice card) has **Copy name**, **Copy ID** and **Project defaults**.

![The More actions menu on the voice card: Copy name, Copy ID and Project defaults](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/voice-card-more-actions-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=e9abba38f7807ddad636e64fb57ead66)

## Expressiveness

**Expressiveness** controls the emotional range and variation in the voice: stable for a consistent sound, expressive for a more dynamic sound.

| Setting | Range | Default | What it changes |
| - | - | - | - |
| **Expressiveness** | 0 (**Stable**) to 2 (**Expressive**), in steps of 0.1 | 1 | Low values keep the delivery even and steady; high values give it more variation |

Expressiveness goes back to 1 each time you open the page or click **Create new**. To hear the difference, generate the same script twice with different values and [compare the takes](https://docs.tryhamsa.com/media/text-to-speech/history.md#compare-two-takes).

## Dictionaries

Dictionaries tell the voice how to say names, brands and terms: each entry is a word and how it should be read. Before your script is read, Hamsa replaces each entry's word with its reading.

**1. Open the list**

In **Dictionaries**, click **Manage**.

**2. Turn dictionaries on**

Select the dictionaries you want for this script. Use "Search dictionaries…" to find one, or **New dictionary** to create one. You can also edit or delete a dictionary from here.

![The Dictionaries dialog with a search field, New dictionary, three dictionaries with checkboxes and their entry counts, Selected 1 of 3 dictionaries and Done](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/manage-dictionaries-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=b291175bce539fd2839b418b0bc6cc07)

**3. Close**

Click **Done**. The section shows the first two dictionaries that are on and a "+N" button that lists the rest. The words they change are highlighted in the script.

You can turn on up to 10 dictionaries for one script. The dictionaries stay on, for any voice, until you reload Hamsa Media. To have them on every time, set them in [Project defaults](https://docs.tryhamsa.com/media/workspace/project-defaults.md).

To add an entry without leaving the script, select the word and use **Add pronunciation**. See [Add a pronunciation from the script](https://docs.tryhamsa.com/media/text-to-speech/generate-speech.md#add-a-pronunciation-from-the-script). For entries, import and export, see [Dictionaries](https://docs.tryhamsa.com/media/voices/dictionaries.md).

> [!WARNING]
>
> An entry changes whole words only, and its letter case must match: an entry for "AI" doesn't change "AIR" or "ai". Arabic harakat are ignored. See [How a dictionary changes the audio](https://docs.tryhamsa.com/media/voices/dictionaries.md#how-a-dictionary-changes-the-audio).

## Speed

There is no speed control in the app: speech is generated at the voice's normal pace. If you remix a job made through the API with another speed (the API takes 0.5 to 2), the remix keeps that speed, and the estimate follows it.

## When you view a past job

When you open a job from [History](https://docs.tryhamsa.com/media/text-to-speech/history.md), the tab shows the settings that job used: its voice, expressiveness and dictionaries. They can't be changed there. Click **Remix** to start a new script from the job, with its voice and expressiveness.

![A past job open: its script read only, the Voice settings tab showing the voice Omar, no dictionaries and Expressiveness 1, Remix and Create new in the header, and the audio in the player](https://mintcdn.com/hamsa-452fa79a/etUfiFCo8p_6XpY_/images/media/text-to-speech/past-job-settings-light.webp?fit=max\&auto=format\&n=etUfiFCo8p_6XpY_\&q=85\&s=beb6e12ffa50c883ffa67ebb41039e05)

If the job's voice no longer exists, the tab shows "Voice not found". Click **Remix** and pick another voice.

## Good to know

* A star (favorite) is yours and per project; it doesn't star the voice for teammates.
* **Recently used** is kept in this browser; it fills again on another computer as you generate.
* If the selected voice is deleted, the page goes back to the default voice.
* On a narrow window or a phone, **Voice settings** is a button in the header that opens the tab in a side panel.

## Related

> [card] [Voices library](https://docs.tryhamsa.com/media/voices/voices-library.md)
>
> Browse, filter and compare voices.

> [card] [Dictionaries](https://docs.tryhamsa.com/media/voices/dictionaries.md)
>
> Create entries, import and export.

> [card] [Generate speech](https://docs.tryhamsa.com/media/text-to-speech/generate-speech.md)
>
> Write the script and generate.

> [card] [Project defaults](https://docs.tryhamsa.com/media/workspace/project-defaults.md)
>
> Start with the same voice and dictionaries.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.