AI Assistants
How does voice cloning and voice processing work with the VOCALIS AI Agent?
Voice library, custom cloning and advanced voice settings.
Your agent's voice is their sonic identity. Choose from an extensive library of professional voices, or create a custom voice from a recording.
Three options available
1. Library of existing voices
Access a wide selection of high-quality pre-trained voices via ElevenLabs and Cartesia. Each voice is available in several tones: formal, casual, warm, authoritative…
2. Custom Voice Cloning
Create a synthetic voice that sounds like a real person speaking. The cloned voice can be used in Pipeline and Dualplex modes.
| Supplier | Requirements for cloning |
|---|---|
| Cartesia | A single audio file, minimum 10 seconds, one speaker only, with no background noise |
| ElevenLabs | Several samples, more than a minute in total, a single speaker, with no background noise |
3. Cartesia Voice Sonic 3 (New)
The Cartesia Sonic 3 TTS engine delivers high-fidelity audio quality with advanced emotion control. It supports voice cloning and SAML tags to adjust pitch, intensity, and expressiveness in real time.
Advanced voice settings
| Setting | Beach | Effect |
|---|---|---|
| Temperature | 0.0 – 1.0 | Lower pitch = stable but less expressive voice. Higher pitch = more dynamic and creative voice. |
| Silence period before hanging up | 30-45 sec | Waiting time if the other party does not answer before ending the call. |
| Maximum call duration | 20 – 1200 sec | Absolute limit on the duration of a call to control costs. |
