Plain-English definitions for the terms that come up when shopping for an AI phone receptionist.
An AI receptionist is software that answers a small business's inbound phone calls in real time, conducts a conversation with the caller, captures structured intake details (name, phone, reason for calling), books appointments to the business's calendar, and texts confirmation to the caller — all without a human picking up the phone.
Read more →HIPAA mode on an AI receptionist is the four-part compliance bundle — Business Associate Agreement (BAA), 6-year transcript retention, explicit per-call recording consent, and audited access controls — that makes the AI legal for healthcare practices handling Protected Health Information (PHI).
Read more →Voice cloning is the AI-receptionist feature that synthesizes the business owner's actual voice (from a short recorded sample) and uses it as the receptionist's voice on every call, so callers hear the owner's voice answering even when the owner isn't there.
Read more →BYO (Bring Your Own) phone number is the model where the customer provisions and owns the phone number under their own Twilio (or equivalent telephony) account, and the AI receptionist vendor merely routes calls through it — versus the vendor providing a number under their own account that the customer loses on cancellation.
Read more →Two-party consent (also called 'all-party consent') is the legal requirement, in 12 US states, that both parties to a phone call must explicitly agree to have the call recorded before any recording can begin — versus the federal default of one-party consent where only one participant needs to agree.
Read more →A Large Language Model (LLM) is the AI system inside an AI receptionist that understands the caller's words, decides what to say, and generates the response in real time. Examples include Anthropic Claude (RingDispatch's choice), OpenAI GPT, Google Gemini, and Meta Llama.
Read more →Text-to-speech (TTS) is the technology that converts the AI receptionist's text response into spoken audio that plays through the call. Modern TTS engines (ElevenLabs Flash, OpenAI TTS, Google Cloud Text-to-Speech) produce voices indistinguishable from human recordings in most accent and emotion ranges.
Read more →Speech-to-text (STT), also called Automatic Speech Recognition (ASR), is the technology that converts the caller's spoken audio into text transcripts that the LLM can read. Deepgram, OpenAI Whisper, and Google Cloud Speech-to-Text are the dominant engines for AI receptionist applications in 2026.
Read more →An intake script is the structured set of questions the AI receptionist asks every caller — name, contact info, reason for calling, vertical-specific details — to capture enough information that the business owner can act on the call without a follow-up. Different verticals require different intake scripts: plumbing emergency intake differs from dental appointment booking differs from legal personal-injury intake.
Read more →After-hours mode is the configurable behavior an AI receptionist follows when a caller dials outside the business's normal hours of operation. Typical settings are 'take a message' (safest), 'emergencies only' (book emergencies + page on-call staff, take a message for routine), and 'always on' (book everything 24/7).
Read more →Dispatch, on an AI receptionist, is the routing of a booked call to a specific team member or vehicle based on the caller's request, the service type, the caller's location, or the urgency. A single-operator business doesn't need dispatch; a multi-staff or multi-location business does.
Read more →Mid-call language switching is the AI-receptionist capability that auto-detects when a caller switches from one language to another mid-conversation (most commonly English to Spanish, but also to Vietnamese, Mandarin, Korean, and 28 others) and continues the conversation in the new language with a native accent, without transferring the call.
Read more →