Customer Support and Voice AI
AI receptionist for small business: calls, bookings and handoffs
What an AI receptionist does on a real call, the rules that apply, a sample greeting and a pass/fail checklist to launch one safely.
WhatsApp Agents
WhatsApp Voice + Text Agent
Voice notes and typed messages on WhatsApp both get understood and answered by the same agent, on the same number.
Many customers would rather send a voice note than type, and plenty of chat tools simply ignore them. This agent listens to voice notes and reads typed messages on the same WhatsApp number, works out what the customer wants, and answers from your approved documents or routes the chat to your team. Customers message the way they like, and nobody repeats themselves.
- Voice notes transcribed and answered, not ignored
- One agent and one number for voice and text
- Transcripts saved with the rest of the conversation
$3,500
20–30 working days
Optional maintenance $249/mo
What it is
WhatsApp Voice + Text Agent is a WhatsApp agent that understands both voice notes and typed messages on one number, then answers or routes the conversation. It does the same support job as WhatsApp AI Support Agent, plus voice-note transcription. Choose it only when a real share of customers send voice notes.
Why businesses buy it
When a customer sends a voice note and gets a reply asking them to type instead, many simply give up. Staff listening to every voice note by hand is slow and easy to fall behind on. This agent treats voice and text the same way, so every message gets a proper answer quickly and your team only steps in when needed. The written scope lists the flows, documents and systems included, so you know exactly what you are getting before you commit.
Problems this solves
Customers send voice notes that staff must stop to listen to
The agent listens to voice notes and understands them just like typed messages. It answers or routes the conversation, so staff are not replaying audio between other tasks.
Voice messages get missed because nobody can skim them
Each voice note is understood and handled in the same flow as text. Requests are answered or routed instead of sitting unheard in a chat thread.
Customers who prefer speaking are asked to type instead
Customers can message however suits them on the same WhatsApp Business number. The agent handles voice and text in one conversation, without asking them to switch.
Mixed voice and text threads reach the wrong person
Using up to five conversation flows, the agent decides whether to answer or route each request. Routing follows the rules agreed in your guardrail configuration.
A good fit if
A meaningful share of your customers send voice notes on WhatsApp instead of typing, especially for longer or more detailed questions.
You want one agent and one number handling both formats, rather than voice notes going unanswered or transcribed by hand.
You already have FAQs, a catalog or booking rules that answer most of what people ask, by voice or by text.
You want a clean handoff to a human agent inbox when either a voice note or a text message needs a person.
Not the right fit if
Almost none of your customers send voice notes, so a text-only build covers the same ground for less. See WhatsApp AI Support Agent.
Your WhatsApp traffic is mostly new leads to qualify and sell to, and few send voice notes. See WhatsApp AI Sales Agent.
The job is bringing lapsing customers back, not answering how they messaged. See WhatsApp AI Retention Agent.
The message is a plain transactional update, not a two-way conversation. See WhatsApp Notification Agent, which sends event-triggered updates and answers simple follow-up replies.
You need phone calls answered, not WhatsApp voice notes. That sits outside this build's scope; use our Custom AI enquiry form for a phone-based project.
How it works
- A customer messages your WhatsApp Business number, as a typed message or a voice note, which opens a session so the agent can reply freely for 24 hours.
- A voice note is transcribed and understood the same way a typed message is, so the customer never has to repeat themselves in the other format.
- The agent searches your approved documents, up to 100 of them, for the answer, whether the question arrived as text or as audio.
- For a booking or a multi-step request, it works through the steps in order, checking each one before moving to the next, regardless of format.
- If the question sits outside the documents, or confidence is low, the chat hands off to your human agent inbox with the transcript and history attached.
- Every conversation, whichever format it started in, the resolution and whether it needed a human are logged to your CRM automatically.
What you provide within 72 hours
Your portal lists these the moment you order. Delivery starts the working day after everything is in and your access is checked.
- Admin access to your WhatsApp Business Platform account and a verified Meta Business account.
- The phone number the agent should use for both voice notes and text.
- Your FAQs, catalog or booking rules covering the questions people ask by either format.
- Approved wording for the message templates Meta needs to review.
- The inbox or person a conversation should reach when it needs a human.
- Admin access and API keys for every connected system, shared by adding our team as a user or creating a key you control, never by email.
- Your business information (company basics, hours and brand voice), shared during intake.
- One decision-maker on your side who can approve the knowledge base and sign off on launch.
- An API key for your own AI model account (for example OpenAI or Google Gemini), created for this build with a spending limit.
The 72-hour rule. Send everything on your checklist, including access, within 72 hours of ordering so your build starts on time.
Timeline
Start
Intake verified and your knowledge base and handoff rules agreed.
Weeks 1–2
Voice transcription, conversation flows and the knowledge base are built.
Week 3
Your CRM and calendar are connected, and templates go to Meta for approval.
Weeks 3–4
40 end-to-end test conversations run, split across voice notes and text, with transcripts sent to you.
Week 4
Your review and two rounds of changes to the answers.
Go-live
Live, with a recorded handoff session, and support starts.
| Option | Delivery time | Price |
|---|---|---|
| Standard | 20–30 working days | $3,500 |
| Express | 15–25 working days | $4,100 |
| Rush | 10–20 working days | $4,750 |
What you receive at handover
- A voice and text agent live on your WhatsApp Business number, connected to up to two systems
- Workflow, prompt and guardrail configuration, including your routing rules
- Answers drawn from up to 100 of your documents or a catalog of up to 500 products
- Up to five conversation flows and five message templates, submitted for Meta approval
- A test report, an admin guide and a recorded handoff session
- 30 days of post-launch support
Expected benefits and how to measure them
- Voice notes can be answered as promptly as typed messages.
- Customers can keep messaging in the way they prefer.
- Staff can spend less time listening back to audio.
- Reply time for voice-note conversations
- Compare first reply times on voice-note threads in your WhatsApp logs, four weeks before and after launch.
- Conversations resolved without staff
- From the agent logs, count voice and text conversations closed without routing, each month.
- Routing accuracy
- Sample routed conversations each month and check that each reached the right person.
Before you start: Count how many voice notes arrive in a typical week and how long they take to answer now. A one-week tally from the inbox is enough.
These are things to track, not promised results. Outcomes depend on your data, your team and your market.
Running costs
The build price is one-time. Running it is paid in your own accounts: WhatsApp conversation fees charged by Meta, voice transcription usage, plus roughly $30 to $150 a month for AI model usage and hosting.
AI usage during the build and testing is also billed to your accounts, so you keep ownership of every provider relationship, including Meta's. Nothing runs through us.
Ask about running costsCustom requirements
If you need something outside the Scope Card, add a note at checkout. You get a written Change Quote, and that work only starts once you accept it. Larger or bespoke projects go through the Custom AI request form instead.
Request a custom versionKeep it running
We recommend monthly maintenance, which covers monitoring, fixes and monthly change hours for things such as a new FAQ or an updated voice greeting. Maintenance starts after the 30 days of included support and can be cancelled any month.
$249/mo (the minimum fee, includes 2 change hours)
Optional. Starts after the post-launch support period. Cancel any month.
How maintenance worksHow this compares
| Do it yourself with a SaaS tool | Hire a custom agency | Upmark Labs | |
|---|---|---|---|
| Price | A WhatsApp chatbot tool charges monthly per agent seat and rarely handles voice notes at all | A custom build gets quoted after discovery calls | Ours is one fixed price, shown on this page before you buy |
| Scope | A text-only bot leaves voice notes unread or ignored | An agency agrees scope during the project | Ours understands voice notes and typed messages on the same number |
| Delivery | A chatbot tool needs weeks of content mapping before it works well | An agency project usually runs several weeks, set per project | Ours has a fixed delivery time shown before you buy, with faster options |
| Handoff | A text-only bot often loses a voice note entirely on handoff | An agency build depends on what was agreed | Ours hands off with a transcript and the full history attached |
Questions about this product
How much does WhatsApp Voice + Text Agent cost?
The price and delivery panel on this page shows one fixed build price for each delivery speed. It covers the full Scope Card: voice note transcription and text handling on one WhatsApp number, up to two connected systems, five conversation flows, five Meta templates, 40 test conversations across both formats and 30 days of support.
How does WhatsApp Voice + Text Agent handle a voice note?
It transcribes the voice note and works out what the customer wants in the same way it would for a typed message, then answers from your approved documents or routes the conversation. The customer never has to repeat themselves in text, and the transcript is saved alongside the rest of the conversation history.
What does WhatsApp Voice + Text Agent cost to run each month?
Running costs include WhatsApp conversation fees charged directly by Meta, voice transcription usage, and roughly $30 to $150 a month for AI model usage and hosting. You pay those providers directly, in your own accounts, so you keep ownership of every relationship. We charge nothing per conversation or per voice note.
Why choose WhatsApp Voice + Text Agent over WhatsApp AI Support Agent?
Choose the Voice + Text Agent when a real share of your customers send voice notes and you want them answered without asking anyone to type. WhatsApp AI Support Agent is the simpler choice when almost everyone types, since it covers the same kind of knowledge base without transcription. The price panels on both pages let you compare them directly.
What do we need to provide before WhatsApp Voice + Text Agent can launch?
We need your WhatsApp Business Platform account, a verified Meta Business account, your FAQs and any booking rules, and the inbox a conversation should hand off to when it needs a person. Account access is shared by adding our team as a user or creating a key you control, never by email, and one person approves decisions.
Will customers know a bot is replying to their voice note?
Yes, the conversation opens with a short disclosure in your own wording, so customers know before they send a voice note or a message. Meta's WhatsApp Business Platform expects clarity about automated messaging. You approve the exact wording during intake, and it can be changed later through monthly maintenance.
Can we update the FAQs or documents WhatsApp Voice + Text Agent uses?
Yes, adding a document or updating an FAQ within the 100-document limit fits inside the monthly change hours of maintenance. Expanding past that limit, or adding a new conversation flow, comes through an add-on, quoted before it starts, without disrupting the live agent.
Guides on this topic
All guides Stop losing whatever arrives as a voice note.
Order WhatsApp Voice + Text Agent today and it goes live on a fixed delivery date. The scope above is exactly what you get, in writing, before you pay.
Works well with
WhatsApp Agents
WhatsApp AI Lead Qualification for Zoho CRM
Qualifies inbound WhatsApp leads and routes them to sales, connected natively to Zoho CRM so it reads and updates leads, contacts, deals…
WhatsApp Agents
WhatsApp AI Lead Qualification for Salesforce
Qualifies inbound WhatsApp leads and routes them to sales, connected natively to Salesforce so it reads and updates leads, contacts, accounts, opportunities,…
WhatsApp Agents
WhatsApp AI Lead Qualification for HubSpot
Qualifies inbound WhatsApp leads and routes them to sales, connected natively to HubSpot so it reads and updates contacts, companies, deals, activities…