Skip to main content
Choosing the right voice is crucial for your agent. The voice sets the tonality, personality, and clarity of every call, whether the agent sounds warm or professional, calm or energetic, and how natural and easy it is to understand. It shapes how callers experience your brand and can make the difference between a conversation that feels human and one that feels robotic. Take time to pick a voice that matches your use case and your audience. Vatel offers a wide variety of voices from various model and TTS providers. You can choose from neural, conversational, and multilingual options so your agent sounds right for your product and region.

​
Why voice matters

The voice you assign to your agent directly affects tone, pace, and expressiveness. A support agent might benefit from a calm, clear voice; a sales or booking agent might use something more upbeat and confident. Consistency between the voice and your prompts (and the rest of your flow) makes the experience feel coherent.
Test different voices with real or sample conversations before going live. The voice picker in agent settings lets you preview and compare options.

​
Providers and categories

Voices are grouped by provider. Each provider offers a distinct set of voices and characteristics. Below is an overview of the main categories and providers available in the Agent Editor.

ElevenLabs

High-quality neural voices with strong expressiveness and natural prosody. Well suited for conversational agents where tone and personality matter. Many voices and languages available.

OpenAI

Text-to-speech voices that are clear and consistent. Good for agents that need reliable, neutral delivery and broad language support.

Cartesia

Low-latency, expressive TTS built for real-time conversation. Strong for voice agents that need natural turn-taking and responsive delivery.

Fish Audio

Realistic, high-fidelity voices with strong multilingual support. Good for agents that need natural-sounding speech across a wide range of languages and accents.

Grok

TTS voices from xAI optimized for clarity and natural pacing. Suitable for agents where clean, direct delivery is the priority.

MiniMax

High-quality multilingual TTS with strong support for Asian languages. Well suited for agents serving Chinese, Japanese, Korean, and other non-English audiences.

Hume AI

Empathic voice synthesis designed to express and respond to emotional nuance. Well suited for support or coaching agents where tone and emotional resonance matter.
Additional providers may be available depending on your plan and region. In the console, open the voice picker in your agent settings to see the full list of providers and voices you can use.