Resources · Glossary
AI and voice terms, explained simply.
A plain-English guide to the words you'll hear when building AI assistants and automating customer conversations.
A
- AI agent
- Software that understands requests, makes decisions and takes actions on its own, such as booking an appointment or issuing a refund.
- AI receptionist
- An AI agent that answers incoming calls, greets callers, answers questions and routes or books on their behalf.
- ASR (Automatic Speech Recognition)
- Technology that converts spoken audio into text. Also called speech-to-text.
B
- Barge-in
- When a caller interrupts the assistant mid-sentence. A good voice agent stops and listens immediately.
C
- Code-switching
- Mixing two languages in one conversation, like Hindi and English in Hinglish.
- Containment rate
- The percentage of conversations fully resolved by the AI without a human stepping in.
- CRM
- Customer Relationship Management software, such as Zoho or HubSpot, used to track customers and deals.
D
- DND (Do Not Disturb)
- TRAI's registry of numbers that have opted out of commercial calls and messages in India.
- DPDP Act
- India's Digital Personal Data Protection Act, 2023, which governs how personal data is collected and used.
H
- Handoff
- Transferring a conversation from the AI to a human agent, ideally with a summary of what happened.
- Hallucination
- When an AI model gives an answer that sounds confident but is incorrect or made up.
I
- IVR (Interactive Voice Response)
- A phone menu system where callers press keys to navigate. Conversational AI is replacing it.
K
- Knowledge base
- The documents, FAQs and data an assistant uses to answer questions accurately.
L
- Latency
- The delay between a caller finishing speaking and the assistant starting to reply. Lower is better.
- LLM (Large Language Model)
- An AI model trained on large amounts of text that can understand and generate language.
O
- Omnichannel
- Serving customers consistently across voice, SMS, WhatsApp, email and web chat.
P
- Prompt
- The instructions given to an AI model that shape how it behaves and responds.
R
- RAG (Retrieval-Augmented Generation)
- A technique where the AI looks up relevant information from a knowledge base before answering.
S
- Sentiment analysis
- Detecting whether a customer sounds happy, neutral or frustrated during a conversation.
T
- TTS (Text-to-Speech)
- Technology that converts written text into spoken audio.
V
- Voice cloning
- Creating a synthetic voice that sounds like a specific person, with their consent.
W
- Webhook
- An automatic message sent from one app to another when something happens, such as a call ending.