What Voice AI Voice Cloning Does
Voice AI voice cloning creates a reusable custom voice from a real speaker’s audio sample. You can create and manage cloned voices directly from the My Voices section inside Voice Selection. Cloned voices can later be assigned to Voice AI Agents.
Audio Sample Requirements
Use a clean recording with one speaker for the best cloning results.
Supported file formats:
MP3
WAV
M4A
OGG
WEBM
MP4
Maximum file size:
10 MB
For best results, avoid:
Background noise
Echo
Multiple speakers
Music
Very low microphone volume
How to Access Voice AI Agents
Go to AI Agents.
Click Voice AI.
Open Agent List.
Open an existing Voice AI Agent or click Create Agent.
How to Open Voice Selection
Inside the Voice AI Agent editor, click the current voice name to open the Voice Selection modal.
How to Start the Voice Cloning Flow
Inside the Voice Selection modal:
Open the My Voices tab.
Click Add Custom Voice.
Select Clone.
How to Create a Cloned Voice
The Clone Custom Voice modal lets you upload or record a source audio sample.
Enter a clear Voice Name.
Select a provider.
Upload an audio file or record directly in your browser.
Confirm you have the rights and consent to clone the speaker’s voice.
Click Save.
Clone Voice Modal Overview
The Clone Custom Voice modal supports both uploaded audio files and browser-based recording. Supported providers include:
ElevenLabs
Cartesia
The modal also includes:
Microphone selection
Recording playback
Pause and resume controls
Consent confirmation
My Voices and the Cloned Voices Library
The My Voices tab stores all custom voices available for Voice AI Agents.
Custom voices are organized into sections:
Imported Voices
Cloned Voices
Imported voices and cloned voices share the same custom voice library.
Imported voices and cloned voices also share a combined limit of 10 custom voices per company.
How to Assign a Cloned Voice to an Agent
After processing is complete, the cloned voice appears inside the Cloned Voices section.
Select the cloned voice.
Click Save Voice.
Save the Voice AI Agent.
Processing, Previewing, and Failed Clones
Cloned voices are not available immediately after submission. After clicking Save, the cloned voice enters a processing state while Kyrios generates the voice. Once processing completes, the voice becomes selectable for Voice AI Agents. If cloning fails, retry using a cleaner source recording. Common causes of failed or low-quality clones include:
Background noise
Echo
Multiple speakers
Unsupported file format
File size above 10 MB
Low microphone quality
Common Issues and Fixes
The cloned voice sounds unclear
Use a cleaner recording with one speaker and minimal background noise.
The cloned voice failed
Retry using a supported file format under 10 MB.
The cloned voice is missing from Voice Selection
Wait for processing to complete before assigning the voice.
The custom voice limit has been reached
Delete unused imported or cloned voices before creating another custom voice.
FAQ
What providers support voice cloning?
You can create cloned voices using:
ElevenLabs
Cartesia
What is the maximum upload size?
Audio samples must be 10 MB or smaller.
Can I record a sample directly inside Kyrios?
Yes. The Clone Custom Voice modal supports browser-based recording.
Do imported and cloned voices share the same limit?
Yes. Imported and cloned voices share a combined limit of 10 custom voices per company.
What happens after I click Save?
The cloned voice enters a processing state and appears under My Voices → Cloned Voices.
What should I do if cloning fails?
Retry using a cleaner recording with one speaker and minimal background noise.
Is consent required before cloning a voice?
Yes. You must confirm that you have permission and consent to clone and use the speaker’s voice.