Four Js

Genero Voice · Modernize

Your business applications,now answering the phone.

Nothing inside them changes. Genero Voice answers callers in 40+ languages from a knowledge base you maintain, and acts only through the endpoints you choose to expose — your models, your keys, your infrastructure if you want it. An add-on to Genero Enterprise.

Logic untouchedYour keys, your infrastructureBuilt by your platform vendor
Use cases

What a call sounds like

Six illustrative calls your application could take, given the endpoints below.

Order statusRetail
Caller

“Where’s replenishment order 48213 for store 22?”

What it takes

One read endpoint on your order-management application — POST, JSON in, JSON out.

What the caller gets

The live status, spoken back, and a summary recorded against the call — instead of a queue for a dispatcher.

Plant floor, hands busyManufacturing
Caller (line supervisor, mid-changeover)

“Do we have enough of part 7741 for the second shift?”

What it takes

A stock-lookup endpoint; a second endpoint if you want the agent to raise a replenishment request — your application applies its own approval rules to it.

What the caller gets

An answer without gloves off or a walk to a terminal.

Citizen permit lineGovernment
Caller (evening, office closed)

“Is my building permit approved?”

What it takes

A status endpoint on your permitting system, keyed on the reference number.

What the caller gets

An answer while the counter is closed, instead of a voicemail in tomorrow’s backlog.

Reporting a claimFinance
Caller

Reports a minor collision.

What it takes

Your own claim-reporting procedure in the knowledge base, plus a create endpoint on the claims application; you write the conditions under which the agent hands over to a human handler.

What the caller gets

A structured first report, taken once, on the record.

Patient line, after hoursHealthcare
Caller

“I need to move Thursday’s appointment.”

What it takes

Availability and booking endpoints on your scheduling system; you instruct the agent to route anything clinical to a person.

What the caller gets

The change made while the desk is closed.

Non-emergency intakePublic Safety
Caller

Reports a blocked road sign.

What it takes

A case-creation endpoint; you declare a dispatcher as a transfer destination and write the conditions for using it into the agent’s instructions.

What the caller gets

A reference number, and a dispatcher who isn’t a switchboard.

How it works

A source of truth you control

Your documents

Procedures, FAQs, policies. Maintained in the platform by your own team, without a release cycle.

Your applications

Live data and actions over REST APIs and webhooks. Logic unchanged.

Genero Voice agent

AnswerActionHandoverRecord

Genero Voice holds no second copy of your application logic. It asks your application and reports back — your logic stays in the one place you own and maintain.

Update the source — the agent follows at the next session.
01

The owner publishes procedure v2

02

The agent works from it

No ticket, no release cycle, no retraining project.

On the call

What it does — and where that comes from

Answers

The knowledge base and instructions you give it — you define what it may say, in your wording.

Acts

Your APIs, live during the conversation — look up a record, create a ticket, update a status. Your application applies its own rules to every write.

Hands over

Your people. The agent transfers the call when its instructions tell it to, and your team can listen in on a live conversation and take over in supported session types.

Leaves a record

A transcript and a summary for every interaction — what was asked, answered and done.

Capabilities at a glance
  • 40+ languages
  • Voice, text chat, camera and screen sharing
  • Inbound calling
  • Transfer to a human, plus live listen and take-over
  • Knowledge base you maintain — a new version applies at the next session
  • Transcript and summary for every interaction
  • Multiple agents, each with its own tool list
  • Self-hosted or cloud — your provider keys
Works with what you run

Genero Voice reaches the systems you run through one contract: an HTTPS endpoint per action, JSON in, JSON out.

Your Genero applications, and the platforms around them — if your service desk lives in ServiceNow or your CRM in Salesforce, they already speak that language. We confirm each target system with you during scoping, telephony included.

REST APIWebhooksWorks with your telephony

Hear it on your own procedures.

Tell us which application should answer the phone. We’ll build a voice agent on the questions your team answers most often — from your own documents, in the language your callers use — and you’ll hear it answer on your material, not on a demo script.

FAQ

Frequently asked questions

Is Genero Voice part of Genero Enterprise?

It’s an add-on to Genero Enterprise. It reaches Genero and non-Genero applications the same way: through the REST endpoints they expose.

Do I have to change my application to use it?

You don’t rewrite it. The agent talks to your application over REST APIs and webhooks, and your business logic applies its own rules to every write. If your application doesn’t expose those endpoints yet, adding them is the integration work of a pilot, and we scope it with you. Genero Enterprise includes a REST/SOAP web-services layer for exactly that kind of work.

Who keeps the agent up to date?

Your team. The knowledge base is built from the documents your own people maintain — publish a new version and the agent works from it at the next session. No ticket, no release cycle.

Where does our data live?

Recordings, transcripts and session data are stored on the storage you configure — your cloud buckets, or your own infrastructure if you self-host the platform. During a call, the caller’s audio and the full knowledge base attached to that agent are sent to the AI providers you selected. Retention and deletion behaviour is scoped per deployment.

Which AI model does it use?

One of the supported providers — OpenAI, Google or AWS. A speech-to-text pipeline configuration uses additional providers for transcription and speech synthesis. You can bring your own provider keys, so you’re not tied to a single model generation.

What happens when a caller needs a person?

The agent transfers the call to a destination you declared, with the reason for the transfer recorded on the session. Your team can also listen in on a live conversation and take over in supported session types.

How do you know who you’re talking to?

You design that, because you own the rule. The agent asks for whatever your procedure requires — a customer number, a case reference — and checks it against your endpoint before it reads anything back. The same applies to call-recording notices: obligations differ by country, so you write the announcement your jurisdictions require into the agent’s instructions.

Hear Genero Voice on your own procedures.