Skip to main content
Famulor offers three engine types for your assistants’ voice stack. They mainly differ by whether text appears as an intermediate step in the flow — and how latency is traded against controlled reasoning.

​
1. Pipeline (classic)

Flow: Speech → text (transcription) → language model → reply text → speech (TTS). This is the classic pipeline: the model reasons over text, which you can trace and tune. That path takes a bit more time, but it shines when logic, longer replies, or heavy automation matter. Typical use cases: detailed explanations, structured workflows, scenarios where you want maximum control over wording and voice.

​
2. Speech-to-Speech

Here the text stage is skipped: the model listens and speaks directly — similar to ChatGPT’s voice mode (the blue circle). It is usually the fastest mode with the lowest perceived wait — great for a snappy outbound caller or short, reactive dialogs.

​
3. Dualplex

Dualplex combines both approaches: the system decides what fits best at each moment — e.g. fast turn-taking where it helps, and more “thinking room” where needed. For most setups, Dualplex is a strong default. If you need more precision, longer answers, or finer control, switch to Pipeline.
For technical detail, latency ranges, and tables, see Assistant modes. Practical tuning tips are in AI assistant best practices.

​
Quick decision