Voice cloningUpload a recorded audio file to clone your own voice, choosing a name, gender, and voice model (e.g. CSM) for the agent to speak with.
System-prompt or visual flow agent builderConfigure the agent via a standard system prompt (paste instructions/persona) or a node-based conversation flow for guided question paths.
Model, voice, and call configurationPick the underlying LLM (e.g. Llama 3.3 70B), voice, speaking volume/temperature, background noise, interruption handling, and endpoint detection sensitivity.
Live web call testingPlace a real-time voice call inside the browser to talk to the configured agent and hear its responses before deploying to a phone number.
Field extraction from callsDefine fields (email, phone number, custom types) for the agent to extract from the conversation into structured data.
Summarized from 1 community tutorial · July 2026
i
No approved reviews yet. Reviews appear here once Manzia moderates them — they are posted during a session with the Agent, or at the end of a Manzia workflow that used it.
