Voices & Languages

Vaaniq runs on a native-audio AI model — it hears and speaks audio directly, so conversations are fast, natural, and interruptible (no robotic speech-to-text → LLM → text-to-speech chain).

You don't pick or tune a model — every agent runs on the same optimized native-audio engine. You just choose the voice and the language the agent speaks.

Languages

Set the agent's primary language so caller speech is transcribed correctly. The agent can still mix in English words naturally. Supported primary languages include:

• Hindi, Gujarati, Marathi, Punjabi, Bengali
• Tamil, Telugu, Kannada, Malayalam
• English (and English-mixed regional speech)

Voices & pacing

Pick a voice that matches the persona from the built-in native-audio voices. You can also tune speaking pace and enable backchannel acknowledgements so the agent feels attentive while the caller speaks.