Voice AI agent for call centres Voice AI agent

A voice agent that picks up on the first ring, understands Uzbek and Russian speech, books appointments and logs the outcome of every call in your CRM. It also dials your customer base: reminders, order confirmations, surveys. Hard calls go to a live operator.

Price
from $7,000
Timeline
from 5 weeks

Price and deadline are fixed in the contract before work starts

What is included

  • 01A call script: what the agent resolves and what it hands to people
  • 02Understanding and speaking Uzbek and Russian
  • 03Inbound calls answered on your own number
  • 04Outbound calls: reminders, surveys, upsells
  • 05Bookings written straight into the schedule
  • 06The outcome of every call in the CRM
  • 07Transfer of hard calls to an operator
  • 08An SMS after the call with the address and time, via Eskiz

Who it is for

  • Clinics and salons where the front desk cannot keep up with the phone
  • Call centres handling routine inbound calls: order status, address, opening hours
  • Services that need to call their base: reminders, confirmations, surveys
  • Businesses that miss calls in the evening, at weekends and at peak hours

Three levels and prices

Pick the level that fits the task and budget. The exact scope is assembled in the configurator and fixed in the contract.

Scenario

One scenario: reminders or inbound calls

$7,000–12,000

≈ UZS 84M · from 5 weeks

Call centre

Inbound and outbound calls, CRM logging, transfer to an operator

$15,000–30,000

≈ UZS 180M · from 10 weeks

Platform

Scenario builder, brand voice, analytics

$35,000–70,000

≈ UZS 420M · from 18 weeks

UZS amounts are indicative, at the Central Bank of Uzbekistan rate on 14.11.2026. Contracts are in USD, or in UZS at the rate on the invoice date.

In detail

What drives the price

Mostly, how many scenarios the agent runs and how many systems surround it. The Scenario tier covers one job: inbound calls or reminders, in Uzbek and Russian. The Call centre tier adds outbound calling, outcomes in the CRM, booking into the schedule and transfer to an operator. The Platform tier is for when scenarios change often: a builder that lets you edit them without a developer, a brand voice, topic analytics and a quality score for every conversation. Connecting your number and PBX, and an SMS after the call, are separate lines in the configurator. Dialogue complexity matters too: confirming a visit is simpler than helping someone choose a service.

How the work runs, week by week

We start by listening: we take recordings of your real calls, pick out the frequent scenarios and decide which one to automate first. Then we write the script: the greeting, the clarifying questions, what the agent does itself and when it calls a person. We build the agent on OpenAI models and run it against your recordings before it ever hears a live customer. We connect your number and PBX, the CRM and the schedule. Launch starts on a share of calls: every conversation is transcribed, and we fix the script wherever the agent stumbled. You see a working version every two weeks, and the agent takes the full flow only after the pilot.

Uzbek and Russian speech: how we test it

In Uzbekistan people often mix Uzbek and Russian in one sentence, use old street names, dictate a phone number in chunks and call from a noisy car. So we test recognition on recordings of your own calls rather than on a demo, separately for each language, and show where the agent makes mistakes. Everything the result depends on — name, phone number, date and time — the agent repeats back and asks to confirm. After the call the customer receives an SMS via Eskiz with the address and time, so the customer spots a recognition slip right away, not on the day of the visit. The agent replies in the language it was spoken to in and switches when the customer does.

Risks and common mistakes

The first mistake is trying to automate everything at once. An agent that runs one scenario very well earns more than one that answers any question poorly. The second is leaving no way out to a person: if the caller asks for an operator or the question falls outside the script, the agent transfers immediately. The third is long pauses: people hang up when a reply lags, so we measure latency during the pilot. The fourth is outbound calling without rules: call at reasonable hours and only people who expect the call. Finally, a call outcome that never reaches the CRM is as good as lost, so every call leaves a trace on the customer record.

What the contract says

2%

for every day we are late

If we are late, your invoice drops by 2% a day, up to 10% of the contract. So we only name deadlines we can keep.

1

company per niche per city

For a year we do not take on your competitor in your city. What we build for you does not end up next door.

10%

refund after an outside review

Bring your own engineer. They check the code against the contract. If a contracted item is missing, we refund 10%.

Questions

How much does a voice AI agent cost?

From $7,000, with a timeline from 5 weeks. That is the Scenario tier: one scenario, such as inbound calls or reminders, in Uzbek and Russian. The Call centre tier with outbound calls, CRM logging and operator transfer, and the Platform tier with a scenario builder, cost more; their prices are in the tier table. The amount is fixed in the contract.

How long does launch take?

One scenario takes from 5 weeks, including a pilot on part of your calls. You see a working version every two weeks. For every day of delay the invoice drops by 2%, up to 10% of the contract.

Do we need our own telephony?

The agent works on your number: we connect it to your PBX or to your carrier’s line, a separate line in the configurator. Customers keep calling the number they know, and your operators stay on the same line to take transfers.

How much does a minute of conversation cost?

A minute is your carrier’s rate plus speech processing on OpenAI models. We calculate the exact cost during the pilot on your real calls and show a forecast for your volume before the agent takes the full flow.

Will the agent understand Uzbek and mixed speech?

It understands and speaks Uzbek and Russian and switches when the caller does. How accurately it recognises your callers is tested before launch on recordings of your calls, separately for each language, and the agent repeats numbers, dates and times back for confirmation.

Won’t customers be annoyed by talking to a robot?

What annoys people is a robot that does not understand and will not let go. Our agent introduces itself, stays on point and transfers to an operator at the first request or when the question gets complex. We measure transfer rates and pauses during the pilot and adjust the script on the facts.

Other solutions

Practice: AI & Data