Skip to main content

​
Agent settings

Agent settings are available in the Settings tab of the agent builder. Open an agent, select Settings, and configure the options that control conversation handling, the agent’s speech, knowledge base usage, and call control. There are a lot of them, so they are split across separate pages — listed below. Only what has to be decided first stays here.

​
Pipeline mode

Choose how the agent processes conversations:
  • STT-LLM-TTS pipeline — configure separate speech-to-text, language model, and text-to-speech options.
  • Realtime model — use a real-time speech-to-speech model for the voice agent.
Pipeline mode toggle: STT-LLM-TTS or Realtime model
Decide this before the rest: some settings do not work in realtime mode. Such an agent has no separate synthesis step and does not fill fields as the conversation goes, so conversation goals, phrase corrections and pause markup have no effect. The toggles still switch on but do nothing — wherever that is the case, the page says so plainly.

​
Where each setting lives

Prompt, models and knowledge

The global prompt and variable substitution in authored text, execution and classifier models, service tier, cache warm-up, knowledge bases, flex mode.

Voice: STT, TTS and VAD

Speech recognition and synthesis, speech detection and interruptions, background audio, turn language detection, the pre-TTS normalizer.

Phrase corrections and pauses

The organization’s replacement dictionary and [pause N] markup: how the agent sounds when it reads numbers and amounts.

Recording and call control

Call recording and its retention, the duration cap, silence reminders, the transfer failure message.

Conversation goals

What has to be established in a conversation, how the platform marks it done, and what the mechanism deliberately does not do.

Analytics, observers and webhooks

Post-call analysis and custom fields, observers with their reactions, post-call webhooks.