> For the complete documentation index, see [llms.txt](https://docs.ixhello.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.ixhello.com/ixhc2/custom-voices-user-guide.md).

# Custom Voices User Guide

### Overview

The Custom Voices feature allows authorized users to create and manage AI-generated voices using natural language prompts. Powered by ElevenLabs, this feature enables you to generate unique voice profiles that can be used in voice-based flows for a more personalized customer experience.

{% hint style="info" %}
Note: Access to Custom Voices requires the Voice Admin role.
{% endhint %}

***

## Prerequisites

Before creating a custom voice, ensure that:

* You have been assigned the Voice Admin role.
* Your environment has available voice quota.
* ElevenLabs integration is enabled for your environment.

***

## Creating a Custom Voice

Follow the steps below to generate and save a custom voice.

#### Step 1: Open Custom Voices

1. Navigate to Custom Voices from the left navigation menu.
2. Verify that you can access the page.

   * If the menu is not visible, contact your administrator to obtain the Voice Admin role.

   <figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FD5APzEzrVaIFYGPXf4sJ%2Fimage.png?alt=media&amp;token=edbdec40-f055-4672-b714-3a08fc2b572e" alt=""><figcaption></figcaption></figure>

***

#### Step 2: Define Voice Characteristics

Enter a prompt describing the voice you want to generate.

**Example Prompt**

```
Warm, intelligent voice. Calm, polished, futuristic, and reassuring with subtle personality.
```

The prompt helps define the tone, style, and personality of the generated voice.

#### Step 3: Enter Preview Text

Provide sample text that will be used to generate voice previews.

You can:

* Use the default preview text.
* Enter your own custom text to better evaluate the generated voice.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FSlfObhEjMZ6mtYQ5CB0T%2Fimage.png?alt=media&amp;token=cca5855e-05fa-43ce-8193-eba683332a02" alt=""><figcaption></figcaption></figure>

***

#### Step 4: Generate Voice Previews

Click **Generate Voice**.

The system will:

1. Send the voice prompt to Eleven Labs.
2. Generate multiple voice variations.
3. Display voice previews for evaluation.

Example preview output:

* Voice 1
* Voice 2
* Voice 3

Use the preview controls to listen to each generated voice.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FzQT7FFJOp43Pgmnl4lr3%2Fimage.png?alt=media&amp;token=f36cfbe4-14c8-422b-ae27-b7610edf4881" alt="" width="562"><figcaption></figcaption></figure>

***

#### Step 5: Select a Voice

1. Listen to all available previews.
2. Choose the voice that best matches your requirements.
3. If none of the generated options are suitable, update the prompt and regenerate new voice previews.

{% hint style="info" %}
**Important:** Voice previews can only be regenerated before the voice is saved.
{% endhint %}

***

#### Step 6: Save the Voice

1. Enter a unique voice name.
2. Click Create Voice.

Once saved, the new voice becomes available within the environment.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FaImx5UxSkMJy6EWmla8r%2Fimage.png?alt=media&amp;token=2e5e9131-c965-4cc5-8dca-7b7aad5cd502" alt="" width="428"><figcaption></figcaption></figure>

***

## Managing Custom Voices

After a voice is created, it appears in the Custom Voices list.

Available actions include:

#### Rename Voice

Update the display name of an existing voice.

#### Preview Voice

Use the **Preview** button to listen to the saved voice at any time.

#### Delete Voice

Remove a voice that is no longer required.

{% hint style="info" %}
Note: Once a voice has been created, its prompt and generated characteristics cannot be edited or regenerated. To create a different version, generate a new voice.
{% endhint %}

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2Fk602X6bC6OIy4MwMX5VO%2Fimage.png?alt=media&amp;token=30c932de-8789-492a-a7d5-0ef6838481a4" alt=""><figcaption></figcaption></figure>

***

## Voice Quota

Each environment has a configurable limit on the number of custom voices that can be created.

#### Example

* Maximum allowed voices: **5**
* Current usage: **4**
* Remaining available slots: **1**

When the quota is reached, new voices cannot be created until an existing voice is deleted or the quota is increased by an administrator.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FK8oCbrsPxhxbExvooaxS%2Fimage.png?alt=media&amp;token=d4789855-2781-4439-9f69-424c74b0411e" alt=""><figcaption></figcaption></figure>

***

## Using a Custom Voice in a Flow

Once a voice is created, it can be assigned to any voice-enabled flow.

### Step 1: Open the Flow

1. Navigate to Flows.
2. Select the desired flow.
3. Click Edit.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FV2SknItugFfjvoQb5EuT%2Fimage.png?alt=media&amp;token=34149ae5-98c8-439e-83e4-132f5cff9ef0" alt=""><figcaption></figcaption></figure>

***

### Step 2: Configure Voice Settings

1. Open Voice Settings.
2. Navigate to Text-to-Speech (TTS) settings.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FCrNMVYZvJ2QKRf5EbAHc%2Fimage.png?alt=media&amp;token=49876361-2007-470d-b5e5-ca71b4165e8f" alt="" width="563"><figcaption></figcaption></figure>

***

### Step 3: Select ElevenLabs Provider

Choose:

* **Provider:** Eleven Labs
* **Model:** Eleven Multilingual Model

{% hint style="info" %}
The custom voice list is displayed only when Eleven Labs is selected as the TTS provider.
{% endhint %}

### Step 4: Choose Your Custom Voice

1. Open the Voice dropdown.
2. Select the custom voice you created.

Example: Release Demo

### Step 5: Preview the Voice

Click **Preview** to hear how the selected voice sounds before applying it to the flow.

This allows you to verify that the voice matches your desired customer experience.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FT4Qos4opqTeOWylf6drz%2Fimage.png?alt=media&amp;token=91587e42-1152-44e1-acf7-28e1385ea7c8" alt=""><figcaption></figcaption></figure>

***

### Step 6: Save the Configuration

1. Click Next.
2. Click Save.

The flow will now use the selected custom voice for Text-to-Speech responses.

***

### Importing Existing ElevenLabs Voices

Organizations may already have custom voices created directly within ElevenLabs.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FeDDiA9coFTj0V0qa6E5t%2Fimage.png?alt=media&amp;token=69c6dcb5-e18f-4382-b3e0-4613a8e1b848" alt="" width="425"><figcaption></figcaption></figure>

At present, imported voices must belong to the same ElevenLabs account configured for the environment.

This means:

Voice IDs created under the configured ElevenLabs account can be used.

Voice IDs from external or personal ElevenLabs accounts cannot currently be imported directly.

The current implementation supports a single configured ElevenLabs account.

<figure><img src="https://4187494610-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FFe4cpY1jy0FJ4kgC7wSx%2Fuploads%2FwqfiVEZSSQLv9OWX6HHK%2Fimage.png?alt=media&amp;token=049cba66-3074-4770-b004-8ef7e6b8f2bd" alt="" width="433"><figcaption></figcaption></figure>

***

### Role-Based Access

| User Role     | Capability                                          |
| ------------- | --------------------------------------------------- |
| Voice Admin   | Create, manage, import, and configure custom voices |
| Standard User | Use configured voices within flows                  |
| Standard User | Cannot create or manage custom voices               |

***

## Best Practices

* Use descriptive prompts that clearly define tone, personality, and speaking style.
* Test multiple voice variations before saving.
* Use meaningful voice names to simplify identification.
* Preview voices within the flow before publishing.
* Regularly review voice usage to stay within configured quotas.

***

## Limitations

* Only users with the Voice Admin role can create and manage custom voices.
* Voice characteristics cannot be modified after creation.
* Voice prompts cannot be regenerated once the voice is saved.
* Voice availability is subject to environment-specific quotas.
* ElevenLabs must be configured as the Text-to-Speech provider to use custom voices.

***

## Conclusion

The Custom Voices feature enables Voice Admin users to create unique AI-generated voices using natural language prompts powered by ElevenLabs. Generated voices can be previewed, saved, managed, and assigned to voice-enabled flows through Text-to-Speech settings, helping organizations deliver more engaging and personalized voice experiences.


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.ixhello.com/ixhc2/custom-voices-user-guide.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.