Overhauling Customer Service with Neural Networks and Advanced Generative Models

Using computer vision, deep learning architecture, proactive monitoring and smart agent assist

Objectives

  • Audio quality issues in customer service calls impact 5-7% of call volume
  • The audio call quality is manually evaluated and scored – overall rating based on any audio quality issues observed in the recorded call
  • The available tools use RMSE and Spectrogram to visualize/analyze signals, however, if audio has echo, lag, low volume, or background noise, it goes unrecognized
  • Missing RCA and proactive monitoring of the problematic calls

Challenges

  • The manual process of evaluating a very large
    number of audio call files is cumbersome and time-consuming
  • The Audio quality issues are not captured and/or scored based on multiple types of audio quality issues
  • There is a need to monitor call quality over regular
    intervals to identify patterns by geography, carrier,
    time of the day, day of the week
  • Support agents need real-time assistance to reduce
    resolution time

Solutions

Leveraged Support Vector Machine (SVM), powerful supervised classification algorithm, to identify the optimal hyperplane that best separates data into distinct classes. The audio quality classification is measured by using dataset including 32 audio metadata attributes and high-dimensional spaces to achieve 97% overall accuracy with testing data.

Built and implemented end-to-end solution for Audio Quality Solution which can:

Google-Audio-Quality-Imge03
Measure multiple Audio Quality attributes for 2-way voice calls between customer service representative and end-customer
Google-Audio-Quality-Imge04
Audio Quality Scoring Framework to measure the customized weighted scoring based on multiple Audio Quality attributes
Google-Audio-Quality-Imge05

Call insights dashboard to map the regular performance and highlight anomalies

Google-Audio-Quality-Imge06
Assist Agents with sentiment analysis, user preferences to give a hyper-personalized customer experience

Outcome

~97% Accuracy
achieved for 6 audio quality attributes
Processed 6 Mn audio calls/month With 5-7% problematic calls

23% improved
Customer Satisfaction
Score

Estimated $5Mn per
annum cost savings
with improved call
quality
50% reduction in processing time
Reduction in support
hours by ~60%
Tech Stack

Audacity, SoundCheck, FastAudio, pyAudioAnalysis, Storage Buckets, Google App Engine, Cloud Composer, TensorFlow, G..K.E

Circle-01

DevOps Engineer

Industry experience: 10 Years
Location: Pune

To apply for this opportunity, write to us at careers@agivant.com with the following subject: AG5111-DevOps Engineer

Agivant is a new-age AI-First Digital and Cloud Engineering services company that drives Agility and Relevance for our client's success.

Powered by cutting-edge technology solutions that enable new business models and revenue streams, we help our customers achieve their trajectory of growth.

Agility is a core muscle, an integral part of the fabric of a modern enterprise.

To succeed in an ever-changing business environment, every modern organization needs to adapt and renew itself quickly. We help foster a more agile approach to business to reconfigure strategy, structure, and processes to achieve more growth and drive greater efficiencies.

Relevance is timeless, and is the only way to survive, and to thrive.

The quest for relevance defines the exponential acceleration of humanity. This has presented us with a slew of opportunities, but also many unprecedented challenges. With technology-led innovation, we help our customers harness these opportunities and address the myriad challenges.