> ## 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.

# Add a voice from a recording

> Clone a voice for your project from an audio file or a recording: what to upload, how to pick the 3 to 20 seconds it is cloned from, and how to preview it.

You can add a voice of your own, for example your brand's voice or a staff member's, from a short recording. The voice is cloned from 3 to 20 seconds of clear speech. Once created it is a project voice: "Everyone in this project can use it."

## Before you start

* You need to be an owner or admin. Adding voices needs the voices create permission, which members don't have by default. See [Roles and permissions](https://docs.tryhamsa.com/overview/projects/roles-and-permissions.md).
* Only clone a voice you have the right to use.
* Have a recording of one person speaking, at least 3 seconds long.

## Add the voice

**1. Open Add voice**

Go to **Voices** and click **Add voice**.

**2. Fill in the voice details**

In **Voice details**, set:

* **Name**: required.
* **Language**: Arabic or English.
* **Dialect**: required for Arabic. For English it is "Not required for English".
* **Tags**: "Select one gender and one style (2 tags total)": **Male** or **Female**, and **Narrator** or **Conversational**.
* **Cover image**: optional. JPG, JPEG or PNG, at most 5 MB.

![The Add voice dialog on step 1, Voice details: name, Arabic with the Saudi Arabia dialect, Female and Conversational selected, the uploaded audio ready to play and its transcription](https://mintcdn.com/hamsa-452fa79a/7GY0-8K2sfzXu6G7/images/agents/voices/add-voice-audio-ready-light.webp?fit=max\&auto=format\&n=7GY0-8K2sfzXu6G7\&q=85\&s=353dcf567da9153b56175ce5a162b5ae)

**3. Add the audio**

Under **Input audio**, choose **Upload** or **Record**.

* **Upload**: MP3, WAV, WebM, OGG, AAC, M4A or FLAC, at most 35 MB.
* **Record**: click **Start recording**, speak, then **Stop recording**. "Record 3 seconds or more; over 20, you pick the part."

**4. Pick the part to clone from, if asked**

If the audio is longer than 20 seconds, **Pick the part to clone from** opens. "A voice is cloned from 3 to 20 seconds: drag the handles or the selection." Click **Play the part** to hear it, then **Use this part**.

![Pick the part to clone from, with a 30-second recording, the first 20 seconds selected between two handles, and Choose another file and Use this part](https://mintcdn.com/hamsa-452fa79a/7GY0-8K2sfzXu6G7/images/agents/voices/add-voice-trimmer-light.webp?fit=max\&auto=format\&n=7GY0-8K2sfzXu6G7\&q=85\&s=803b80e04ac078030a5d9965b08da0b6)

**5. Confirm the transcription**

The **Transcription** fills in on its own from the audio. Read it, correct it if needed, and click **Confirm transcription**. You can't continue until it is confirmed.

**6. Preview the voice (optional)**

Click **Next**. In **Generate custom template**, type at least 5 words under **Text** and click **Generate** to hear how the new voice sounds.

**7. Create it**

Click **Create**. "Voice created successfully". The voice appears under **Project voices**.

## Audio limits

| Item | Limit |
| - | - |
| Audio formats | MP3, WAV, WebM, OGG, AAC, M4A, FLAC |
| Audio file size | At most 35 MB |
| Shortest audio | 3 seconds |
| Part used for cloning | 3 to 20 seconds, picked in the trimmer when the audio is longer |
| Longest file the trimmer reads | 10 minutes. "Choose a file up to 10 minutes, then pick the part to use." |
| Longest recording | 10 minutes, then it stops on its own |
| Cover image | JPG, JPEG or PNG, at most 5 MB |

## Choose the style

* Pick **Conversational** for a natural sound "with natural pauses, breathing, and expressions", or **Narrator** for "fast-paced, continuous delivery that's clear, direct".
* Preview the voice with a sentence your agent will really say.

## After you add it

* Use it in an agent like any other voice. See [Choose a voice](https://docs.tryhamsa.com/agents/voices/choose-a-voice.md).
* A voice can't be edited after it is created. To change it, add it again and delete the old one (row menu > **Delete \{name}**).
* Deleting a voice can't be undone.

## Related

> [card] [Choose a voice](https://docs.tryhamsa.com/agents/voices/choose-a-voice.md)
>
> Browse voices and pick one for an agent.

> [card] [Pronunciation dictionaries](https://docs.tryhamsa.com/agents/voices/dictionaries.md)
>
> Fix how the voice says specific words.


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