Skip to main content
Agent
An AI-powered voice assistant built on Atoms that handles phone calls or web conversations.
API Call
A request made by your agent to an external service during a conversation to fetch or send data.
Audience
A contact list used for outbound campaigns.
Branch
A connection between nodes in Conversational Flow that determines conversation path based on conditions.
Campaign
An outbound calling program that contacts a list of phone numbers using your agent.
Cold Transfer
A call transfer where the receiving party is connected immediately without any context or briefing.
Condition
A rule in Conversational Flow that determines which branch to follow based on user response.
Conversational Flow
An agent type that uses a visual workflow builder for structured conversations.
Denoising
Audio processing that removes background noise from caller audio for clearer speech recognition.
Disposition
The outcome category of a call (e.g., successful, unsuccessful, transferred).
Electron
Smallest’ voice-optimized Small Language Model (SLM) that handles reasoning and response generation with sub-500ms latency.
Emotive Model
A speech-to-speech AI model with emotional tone and natural expression (Beta).
End Call
The function that terminates a conversation based on defined conditions.
Global KB
A Knowledge Base shared across multiple agents.
Inbound Call
A call initiated by a customer to your agent’s phone number.
Knowledge Base (KB)
A repository of documents and information that agents can reference during conversations.
Latency
The delay between when a user finishes speaking and when the agent starts responding.
Lightning
Smallest’ Text-to-Speech model that synthesizes studio-quality voice at 44kHz with 175ms latency. Supports voice cloning and multiple languages.
LLM (Large Language Model)
The AI model that understands intent and generates responses.
Local KB
A Knowledge Base attached to and used by only one agent.
Lock Agent
A feature that prevents edits to an agent’s configuration.
Node
A single step in a Conversational Flow workflow (e.g., Prompt, Transfer, API Call, End).
Outbound Call
A call initiated by your agent to a customer (via campaigns).
PII Redaction
Automatic removal or masking of personally identifiable information from transcripts.
Pulse
Smallest’ Speech-to-Text model with 64ms time-to-first-transcript latency. Supports 32 languages with automatic detection and industry-leading accuracy.
Post-Call Metrics
Analytics and data extracted from completed calls.
Pre-Call API
An API call that executes before the conversation starts.
Post-Call API
An API call that executes after the conversation ends.
Single Prompt
An agent type using one comprehensive prompt for flexible conversations.
SLM (Small Language Model)
Atoms’ optimized models that deliver fast, efficient voice AI.
Smart Turn Detection
Intelligent handling of when the agent should start speaking.
STT (Speech-to-Text)
Technology that converts spoken audio into text.
Transfer Call
Handing a conversation off to a human agent or another phone number.
TTS (Text-to-Speech)
Technology that converts text responses into spoken audio.
Turn Detection
The system that determines when a speaker has finished and it’s time to respond.
Variable
Dynamic values that can be used in prompts (e.g., {{customer_name}}).
Warm Transfer
A call transfer where the AI provides context to the receiving party before connecting.
Webhook
An HTTP callback that sends data when specific events occur.
Whisper Message
A message spoken only to the transfer recipient during warm transfer.
Widget
An embeddable voice interface for websites.
What’s Next
