Cloud infrastructure and voice analytics go hand in hand

Use case overview

Our client, a French payment service provider, wanted to build an open, stable and sustainable solution that will let them use the advantages of the latest DevOps technologies and connectivity options. Their infrastructure was fully in physical data centers and this caused lots of pain - legacy issues, time consuming configuration, no scalability. Of course it goes without saying that all of this resulted in losing revenue.

We had to use our AWS expertise in order to create an architecture design that is resilient and easily scalable. Another main part of the project is using the concept of VoIP, SIP and of course voice analytics. These technologies allow us to drastically improve the quality of the calls, use the unlimited toll-free calling instead of per-minute charges and turn text to speech to get real-time ratings.

Main challenges

Tech stack

Solution

Results

Business pain & challenges

No option for using multiple operators

Not able to link voice bot solutions to the infrastructure

Can’t use the advantages of SIP

Physical data center

Bad voice quality

No flexibility

Tech stack

Voice device

CSR 1000V - Used as a Session Border Controller to handle and route VoIP traffic

AWS

VPC - Virtual Data Center for multi-region architecture
S3 - Using this storage service to save the call records
Glacier - Allows us to keep all the logs archive
Direct Connect Gateway & Direct Connect - Consistent links for connecting Telecom providers
Amazon Chime Voice Connector - Used for recording calls
Amazon Transcribe - Real time speech to text conversion
Amazon Comprehend - Analysing every call to provide automated call ratings

Logs management

Cisco CDR - A standard file format used to get calls details and consolidated into a SIEM
SIPREC - Recording feature to get a full call records for post processing
Splunk - Consolidate logs and create dashboards to visualise statistics

Automation

Terraform - Used to automate the deployments of infrastructure stacks
Ansible - Used to write playbooks that configure Cisco device

Solution

Connect Telecom providers

Carrying customer calls via highly available and consistent links using Direct Connect and Direct Connect Gateway

Handle and manage calls

Create a call routing process in a secure and efficient way

Deliver calls

Forward calls to the client’s apps where the core business logic is handled

Logs management and calls analysis

Process call logs to create dashboards and use a dedicated pipeline for call ratings

Expected results & highlights

Faster ROI

Better connectivity and quality of sound

Better monitoring

Less expenses for hardware and calls

Pay-as-you-go, easily scalable infrastructure

Delivery model

Dedicated consultant, who is an AWS expert, working closely with the client’s team, ensuring on-time daily communication

Waterfall methodology & rituals

Weekly status meeting with the client
Constant communication and synchronisation

Roles

AWS Architect / Solution Architect
Network Engineer

Cloud infrastructure and voice analytics go hand in hand

Success story

Use case overview

Main challenges

Tech stack

Solution

Results

Business pain & challenges

Tech stack

Voice device

AWS

Logs management

Automation

Solution

Connect Telecom providers

Handle and manage calls

Deliver calls

Logs management and calls analysis

Expected results & highlights

Delivery model

Waterfall methodology & rituals

Roles

Let's talk

Technofy Ltd.