AI phone assistant for restaurants and hotels
The phone rings hardest when nobody can leave the floor. Tel-Agent takes the booking, the party size and the allergy during service.
- Party size
- Allergies
- Service-hour calls
Why a restaurant phone rings loudest during service
A restaurant's phone competes with a full room. It rings hardest at exactly the hour every person in the building has both hands full, and most of those calls are bookings - which means a call that rings out is a table that goes to the restaurant next door.
How the AI receptionist handles a restaurant booking call
It answers during service without anybody leaving the floor.
It takes the name, the party size, the date and the time.
It asks about allergies and dietary needs, every time.
An allergy, a large group or an accessibility request is flagged so a person sees it before service.
The booking lands written down, and the caller gets a confirmation instead of a busy tone.
What the agent is told to ask
Written by you, in your own words, as the assistant’s instructions.
What a restaurant phone assistant should and should not do
What it does
- Answers through service and after closing
- Takes bookings, changes and cancellations
- Asks about allergies on every call
- Flags groups and access needs for a person to confirm
- Keeps the book on the restaurant's own machine
What it never does
- Confirm that a dish is free from an allergen
- Overbook or promise a table that is not free
- Negotiate on a deposit or a cancellation charge
- Take a card number over the phone
- Send guest data to a third-party booking platform
Caterers
Date, headcount and venue - taken while the kitchen is working.
- Date and headcount
- Venue
- Dietary requirements
Hotels
Answer reception's phone while the desk is dealing with a guest.
- Dates and nights
- Late arrival
- Access needs
Restaurants
Take the booking mid-service, without stopping anybody on the floor.
- Covers and time
- Allergies
- Which sitting
Questions
Why not just let it ring during service?
Because most of those calls are bookings, and a booking that rings out goes to the restaurant next door. The phone is loudest at exactly the hour every person in the building has both hands full.
What about allergies?
An allergy is not a note on a booking - it is the reason a booking needs a person to see it. It can be set to hand over rather than be handled, so it reaches the kitchen rather than sitting in a list.
Does it work for a hotel as well as a restaurant?
The same shape: take the booking, the dates and the party, and transfer anything unusual. A hotel's version of unusual is usually a group, a late arrival or an accessibility requirement.
Can it take a cancellation in time to re-sell the table?
Yes, and cancellations taken early are worth more than the booking they replace - a table given up in the afternoon can still be re-sold for the evening.
Does it handle group bookings?
It takes them and flags them. A large group changes the layout, the kitchen and sometimes the deposit, so it is set to reach a person rather than to be confirmed automatically.
What about hotel enquiries as well as tables?
The same shape with different fields: dates and nights instead of a sitting, and a late arrival flagged the way an allergy is. Both are cases where a written note reaches the right person before service.
Does the EU AI Act allow an AI to answer a booking enquiry?
Yes, under Article 50 of Regulation 2024/1689, which asks that the caller be told they are speaking to a machine - that line belongs in the greeting you configure. Nothing about this work falls under Annex III, the list of high-risk uses. The distinction the Act draws is between assisting and deciding, and this agent assists: it never decides who gets a table or a room. Austria's KI-Servicestelle at RTR publishes the same reading of the transparency duty.
Who is the data controller for what a caller to a hospitality business says on the phone?
You are. A booking enquiry is ordinary personal data, plus whatever a guest volunteers about an allergy, and Tel-Agent is self-hosted with your own model keys, so the recording, the transcript and the record it writes stay on your server and never reach Dpro GmbH. There is no processor agreement to sign with us because we never receive the data, and the source is AGPL-3.0, so you can check that rather than take our word for it.
Can a hospitality business run this in the United States?
There is no single federal AI statute to comply with, but a phone line carries older rules. The TCPA governs automated outbound calls, and consent to record is set state by state. Consent to record is set state by state, and California and Illinois among others want every party to agree. Running the software yourself makes you the caller of record, so the disclosure and the recording switch are yours to set.
Not usable yet. The day it answers a real call will be a release, not a newsletter.
Watch the repository
