Why Create Custom Voices?
Using your own voice makes content unique. It builds a strong connection with your audience. Custom voices are perfect for all your audio needs. They enhance AI dubbing for videos and more.
How to Create a Voice by Uploading an Audio File?
The Create Voice feature lets you generate a custom AI voice by uploading a high-quality audio sample. Follow the guidelines below to ensure the best results.
Creating a Voice
Click the Create Voice button at the top-right corner of the page.
You may see the following creation options:
Upload a Voice Sample
Use this option to create a voice by uploading an existing voice recording/sample.
Step 1:
Click Create Voice
Select Upload a voice sample
Supported File Requirements
To maintain high-quality voice generation, your audio file must meet the following criteria:
File formats supported:
.wav,.flac,.aac,.mp3Audio length: Between 3 seconds and 10 minutes
A minimum of 3 seconds is required to ensure optimal voice accuracy.
Maximum file size: 40 MB
Playback limit: Up to 10 minutes
Audio Quality Guidelines
For best results, make sure the audio clip:
Is clear and high quality
Contains minimal background noise
Has no interruptions, distortions, or overlapping voices
Background Noise Option
If your audio includes background noise, enable the “Audio contains background noise” checkbox.
This helps the system process the audio more accurately.
Step 2: Proceed
Once the file is uploaded and settings are selected, click Next to continue creating your custom voice.
Step 3: Add Voice Details 📝
After uploading your audio file, the next step is to define your voice settings. These details help personalize and accurately categorize your AI voice.
Voice Name: Enter a name for your voice (for example, Female Voice). This makes it easy to identify and reuse the voice later.
Language: Select the language and region that best match the uploaded audio, such as English (United States). Choosing the correct language ensures better pronunciation and natural speech output.
Gender: Choose the gender that aligns with the voice in the audio sample (e.g., Female).
Age: Enter the approximate age of the speaker (for example, 28). This helps fine-tune the voice characteristics and tone.
Select Category: Select a content or voice category.
Voice Description: Add a short description to define the voice style or mood, such as “Narration in soft tone.” This is useful for clarity and future reference.
Step 4: Generate the Voice 🎙️
Once all details are filled in:
Click Back to review previous steps, or
Click Generate to create your AI voice using the provided details.
Click Confirm.
Note: After creation, your voice is saved in My Voices section of the Voice Library for easy access and reuse.
How to Create A Voice From Text?
Text to Voice Synthesis: Craft your perfect voice with a detailed description.
You can describe a voice to create it.
Go to the "Custom Voices" area.
Select the "Create Voice" option.
Type a detailed description of the voice you want. Describe its gender, age, and tone.
It will generate a voice based on your words.
You can then customize this voice further.
Example :
Voice Description: (Describe your voice)
Warm, calm, and expressive, with gentle pacing and soft inflections that create a soothing, immersive storytelling tone.
Text Preview: (Voice Script)
As the sun dipped below the horizon, the quiet village of Elmsworth held its breath. Every evening, just as twilight touched the rooftops, a soft bell would chime from the old clocktower—though no one had lived there for years. Young Clara, curious and brave, stepped into the shadows, determined to uncover the secret that had slept for generations.
Once you click on “Generate Voice”, it will give you two different voices based on your voice description.
Select one of the two voices, then click 'Continue'.
OR click 'Generate' to create a new Text-to-Voice synthesis.
Enter the Name, Language, Gender, Age, Category and Description.
Click 'Save'.
Click Confirm.
What “Generate Random Voice” does?
A pool of preloaded sample sentences.
The system picks a voice description at random from the available set.
Based on the randomly chosen description, the system pulls a suitable TTS voice from the catalog.
The chosen random text is read aloud in the selected random voice.
You hear a preview right away, without setting anything manually.
How Do You Manage Your New Voices?
After creating a voice, you decide its use.
You can keep your voice private. This means only you can access it.
Or, you can publish it to the Marketplace. Sharing it on the Marketplace makes it public. Others can then use your custom voice.
How do I delete a voice from the voice library?
You can delete voices that appear under the Custom and Shared with me sections of the voice library.
Voices listed under Explore and Workspace Voices cannot be deleted.
Congrats!🎉 Now you are able to save your voices and use them freely in your projects. If you need information on how to use a voice form Library in your projects, please refer to this article.