Core System Architecture

Every ANET call involves three coordinated systems.

STT:

• Converts caller speech into transcript text

• Detects pauses and end-of-turn signals

• Provides confidence signals

• Supports language detection

Accurate transcription directly affects intent identification.

The LLM:

• Interprets the caller meaning

• Identifies likely intent

• Determines the next best question

• Applies configured routing logic

• Generates structured summaries

The LLM does not operate independently. It follows configured intent and action rules.

TTS:

• Converts system responses into audio

• Maintains language alignment with STT

• Produces real-time conversational responses

When language changes, the entire stack (STT, LLM, TTS) changes together.