• Home
  • About
  • Our Services
    • AI Applied
    • AI Accelerate
  • Key Industries
    • Finance and Banking
    • Healthcare
    • Manufacturing
    • Retail
    • Product Design and Development
    • Smart City and Infrastructure
  • Resources
    • Case Studies
    • Ebook: How to Bring AI to Your Organization
    • Free Guide: Discussion Questions for AI Readiness
    • New Research: 50 AI Examples from the Fortune 500
  • Projects
    • Unredactor
    • Coronavirus Facts & Myths
  • Coronavirus + AI
    • Coronavirus Facts & Myths
  • Blog
  • Contact us
Manceps

OUR LATEST ARTICLES

Case Study: Using Natural Language Processing to Build a Sophisticated Interactive Voice Response System

Case Study: Using Natural Language Processing to Build a Sophisticated Interactive Voice Response System

Executive Summary

 

A  SaaS company recently engaged Manceps to develop and deploy an AI-powered Interactive Voice Response (IVR) System to help them improve the speed and accuracy of their customer service phone system. By deploying an end-to-end solution that integrated with existing datasets, the organization was able to offer its customers an extremely powerful AI-powered agent who could answer all sorts of questions, troubleshoot their issues, and otherwise anticipate their unique needs.

 

The Problem

 

Despite sophisticated advancements in smart assistants like Apple’s Siri and Amazon’s Alexa, interactive service agents have yet to provide a customer experience that doesn’t leave most users frustrated. To be sure, automation can save customer support teams significant capital; however, most organizations have to balance the cost savings with the inevitable degradation of the customer experience.

 

When our client came to us, their existing phone system connected with customers to an endless forest of menu options that increasingly failed to provide the type of positive customer service their paying subscribers had come to expect. Furthermore, a lack of nuance in their system meant that once their customers finally reached a live agent, the phone tree had failed to sort them to the correct department.

 

Special Requirements

 

Unstructured Documentation

A successful interactive agent needed to be able to tap into a variety of unstructured documents such as knowledgebase articles or FAQs in order to deliver the customer exactly the right information they need. The solution we built needed to seamlessly blend these resources with the agent to prevent the need for complex and ongoing integrations.

 

Sentiment Analysis

The client needed our solution to easily surface frustrated customers in order to provide them with a higher-touch experience and reduce churn.

 

Analytics

Managers needed a way to assess the success and efficiency of their customer service teams without listening to hours of customer interactions or relying upon the customers to participate in post-call surveys, a rare occurrence.

 

Integration

This new system needed to play nicely with their chosen VOIP service. Additionally, they wanted the ability to roll-out these interactive capabilities to other points of customer interaction such as their on-site chatbot and email teams. 

 

International Users

The system needed to leverage the very best in natural language processing as many of their callers had heavy accents that thwarted their existing systems. 

 

Our Solution

After a detailed discovery effort, our team went to work building a state of the art contact center tool that allowed the organization to dramatically modernize their Interactive Voice Response system. This system is now accessible to customers 24/7, answering complex questions with the ease and interactivity of a human. During business hours, the IVR helps live agents better predict and respond to customer needs and delivers ongoing insights to stakeholders via a robust analytics system.
 
At the beginning of this project, we set a natural speech processor to analyze historical audio to uncover insights about topics and trends in customer actions. This ensured that we were able to tailor the agent to the client’s precise requirements and also to identify opportunities to improve the interactive agent and refine their existing knowledge base content.
 
Once this research effort was complete, we began building out the IVR. A series of deeply trained AI models gave the agent the ability to understand complex speech, access unstructured documentation, escalate frustrated customers to human agents, and give stakeholders ongoing visibility into the operational efficiency of the entire department.
 

Results

The interactive agent has already proven invaluable to our SaaS company partners. So far, they have reported a number of impressive improvements as a direct result of this investment. The organization reduced its churn rate, lowered the average time their customers spend on hold, and have seen far fewer escalations from frustrated callers.
 

30.04.2020

317085810024352-photo-1580735995239-eab9cbde7ed6.jpeg

The Complete Guide to Bringing AI to Your Organization

GET THE EBOOK ▾

Get notified when we publish a new story.

Our Most Recent Articles

DevFest West Coast 2020

DevFest West Coast 2020

Video: Machine Learning Engineering with Tensorflow Extended

Video: Machine Learning Engineering with Tensorflow Extended

Video: How to Build a Reproducible ML Pipeline

Video: How to Build a Reproducible ML Pipeline

Video: ML adventures with AutoML and TFHub

Video: ML adventures with AutoML and TFHub

Load More

30232092-r1005-5-15841303371315.jpg

50 AI Secrets: How Every Fortune 50 Company is Using AI Right Now

GET THE REPORT →
OUR LATEST RESOURCES
OUR LATEST ARTICLES

DevFest West Coast 2020

Watch videos of some of the world's top AI experts discuss everything from Tensorflow Extended to Kubernetes to AutoML to Coral.

Video: Machine Learning Engineering with Tensorflow Extended

In this talk, Hannes is providing insights into Machine Learning Engineering with TensorFlow Extended (TFX). He introduces how TFX for machine learning pipeline tasks and how to orchestrate entire ML pipelines with TFX. The audience learns how to run ML production pipelines with Kubeflow Pipelines, and therefore, free the data scientist's time from maintaining production machine learning models.

Video: How to Build a Reproducible ML Pipeline

Solving a data science problem usually requires multiple steps. These steps can include extracting and transforming data, training a model, and deploying the model into production. In this session, we'll discuss how to specify those steps with Python into an ML pipeline. We'll show how to create a Kubeflow Pipeline, a component of the Kubeflow open-source project. The audience will learn about how to integrate TensorFlow Extended components into the pipeline, and how to deploy the pipeline to the hosted Cloud AI Pipelines environment on Google Cloud. The key takeaway is how to improve reuse and reproducibility of the machine learning process.

LOAD MORE

OUR HEADQUARTERS
Headquartered in the heart of Portland, Oregon, our satellite offices span North America, Europe, the Middle East, and Africa.

(503) 922-1164

Our address is
US Custom House
220 NW 8th Ave
Portland, OR 97209

Copyright © 2019 Manceps