Enterprise AICC

Keystone

The full enterprise AICC for finance and the public sector

From call bot and chatbot dialog engines to an agent advisor, text analytics (TA), quality assurance (QA/QC), and knowledge management (KMS) — everything a large finance or public-sector contact center needs, on a single platform.

EnterpriseFinance · Public sectorOn-prem · CCaaS
Contact usAll products
Keystone product screen
OVERVIEW

Keystone transforms legacy contact centers into a modern AICC — CCaaS or on-premises. From call bot and chatbot dialog engines to a real-time agent advisor, text analytics (TA), quality assurance (QA/QC), and knowledge management (KMS), it delivers everything a large finance or public-sector contact center needs on a single platform.

Request a demo
WHO IT'S FOR

Built for teams like yours

"Our legacy contact center has hit its limits"

Aging PBX and IVR systems can't support AI-driven service. Legacy integration APIs let you move to a modern AICC (CCaaS).

"We need large-scale, high-security operations"

For finance and the public sector, security and scale are everything. Deploy safely at scale, on-premises or as CCaaS.

"We want to standardize service quality"

Response quality varies widely between agents. The advisor, QA, TA, and KMS standardize response knowledge and quality.

MODULES

One platform, five core modules

From the dialog engine to knowledge, analytics, and quality — everything a large contact center needs, fully integrated.

CE

Dialog engine

Design call bots and chatbots as node workflows, powered by sLLM/LLM and RAG.

ADV

Agent advisor

Auto-classifies customer intent, extracts keywords, surfaces KMS knowledge, and requests supervisor coaching.

KMS

Knowledge management (AICM)

Structures, searches, and compares FAQs and manuals for accurate answers with no waiting.

TA

Text analytics

Analyzes call volumes, call categories, and top-10 trending and high-frequency keywords to deliver insights.

QA/QC

Quality assurance

Standardizes quality with automated scoring of every call, statistical and comparative analysis, appeals, and re-evaluation.

KEY FEATURES

Key features

FEATURE

Dialog engine · Agent builder

Design call bots and chatbots as node-based workflows, and connect sLLM/LLM to freely compose RAG, conditional branching, external integrations, and agent handoff.

Dialog engine and agent builder
FEATURE

Agent advisor

During the call, it auto-classifies customer intent, extracts keywords, and surfaces KMS response knowledge in real time. In-call supervisor coaching requests, automatic call summaries, and legacy integration APIs — all on one screen.

Agent advisor
FEATURE

TA · Text analytics

Analyze call categories, top-10 trending and high-frequency keywords, and issue-call trends on a period-based call dashboard. Insights like "calls about critical-illness coverage up 300%" arrive as automated reports.

TA text analytics
FEATURE

QA/QC · Quality assurance

Build scoring criteria, scorecards, and evaluation plans with automated and manual templates, then score every call automatically. Run statistical and comparative analysis plus appeals and re-evaluation (up to 3 rounds) with SSO and role-based permissions.

QA/QC quality assurance
FEATURE

KMS · Knowledge management

Structure FAQs and manuals with document templates, and manage them with advanced search, document view/compare, categories, permissions, and activity history. Accurate AI knowledge service with no waiting.

KMS knowledge management
PRODUCT TOUR

More screens

Workflow node detail
Workflow node detailDesign dialog flows precisely, node by node — RAG, LLM, handoff, and more.
Call history detail
Call history detailBrowse the agent advisor's call history and summaries in detail.
TA call analysis
TA call analysisAnalyze the transcript, sentiment, and keywords of each call.
QA comparative analysis
QA comparative analysisCompare quality across agents and time periods.
KMS unified search
KMS unified searchSearch and compare documents, FAQs, and manuals in one place.
Agent monitoring
Agent monitoringMonitor deployed agents' behavior and throughput in real time.
SPECIFICATION

Key specifications

✓
TargetLarge, high-security contact centers in finance, the public sector, and beyond
✓
ModulesDialog engine (CE) · Agent advisor (ADV) · Knowledge management (KMS) · Text analytics (TA) · Quality assurance (QA/QC)
✓
Dialog engineCall bot and chatbot node workflows · sLLM/LLM · RAG and external integrations
✓
Text analytics (TA)Call categories · Top-10 trending/high-frequency keywords · Automated issue-call trend reports
✓
Quality (QA)Automated scoring of every call · Statistical/comparative analysis · Appeals and re-evaluation (up to 3 rounds)
✓
Security & accessSSO · Role-based access control (QA lead, evaluator, trainer, and more)
✓
IntegrationIntegration APIs for legacy PBX and IVR
✓
Deployment & voiceOn-premises / CCaaS · Timbel's own HAIV STT·TTS
CASE STUDIES

Customer stories

The technology behind Keystone is proven in the field

Kakao Pay AICC integrated center STT/TTS
Kakao Pay
AICC integrated center STT/TTS
National Health Insurance Service phone infrastructure rebuild STT
National Health Insurance Service
Phone infrastructure rebuild STT
Daegu Metropolitan City 120 AICC phase 2 build
Daegu Metropolitan City
120 AICC phase 2 build
POBA AI smart contact center build
POBA
AI smart contact center build
View all case studies
FAQ

Frequently asked questions

How is Keystone different from cxVisor?
cxVisor is an AI-native contact center for small and mid-sized teams that need fast deployment, while Keystone is a full enterprise AICC built for large, high-security contact centers in finance and the public sector. It is designed for environments with demanding scale and security requirements.
What modules does it include?
Keystone comprises a call bot and chatbot dialog engine (CE), agent advisor (ADV), text analytics (TA), quality assurance (QA/QC), knowledge management (KMS), and an agent builder. Run the modules you need on a single platform.
Can it be deployed on-premises?
Keystone supports both on-premises and CCaaS deployments, with integration APIs for legacy PBX and IVR systems. Voice is powered by Timbel's own HAIV STT and TTS engines.
What insights does TA (text analytics) provide?
TA analyzes call categories, trending and high-frequency keywords, and issue-call trends to catch changes early — such as a sudden spike in a specific inquiry type — and delivers automated reports with insights for improving responses and staffing.
Can QA evaluate every call, not just samples?
Yes. Automated evaluation scores every call, with statistical and comparative analysis plus an appeal and re-evaluation process (up to 3 rounds). You get full-coverage quality management instead of sampling.
How does it reduce variance between agents?
The agent advisor analyzes customer intent during the call and surfaces real-time response knowledge and coaching from the KMS. Using the advisor, QA, TA, and KMS together lets you standardize response knowledge and quality across agents.
How are permissions and security handled?
Keystone provides SSO and role-based access control, and an on-premises deployment keeps all data processing inside your internal network. It can be operated to meet the high-security requirements of finance and the public sector.
RELATED

Related products

cxVisorWeb AI voice & chat support→HAIV STT·TTSSpeech recognition & synthesis engines→TimbloAI meeting minutes→

Bring Keystone to your contact center

From consultation to tailored deployment, Timbel is with you all the way.

Contact usView all products