Harnessing Scalable Compute and LLMs for Call Center Performance Analytics

Abstract: 

Call centers generate vast amounts of conversational audio data, traditionally difficult to analyze at scale due to the time and effort required for manual review. This session presents an innovative approach leveraging scalable compute and Large Language Models (LLMs) to rapidly and effectively evaluate call center performance.

We demonstrate an automated pipeline where audio recordings from a synthetic bank call center dataset are transcribed into text using Whisper, an open-source speech-to-text model accessed via OpenAI's API. The resulting transcripts are then analyzed by LLMs, such as GPT-4, to summarize interactions and evaluate agent performance across key metrics, including adherence to scripts, problem resolution effectiveness, and customer sentiment.

The session highlights how these AI-driven techniques allow businesses to objectively assess agent interactions, quickly identify training opportunities, and measure customer satisfaction at an unprecedented scale. Additionally, we explore how the Zerve platform's scalable cloud orchestration supports parallel processing, enabling analysis of thousands of audio hours efficiently and cost-effectively.

Attendees will learn:

How to construct an end-to-end analytics pipeline using Whisper and LLMs to turn audio recordings into actionable insights, greatly enhancing operational decision-making.

Techniques for scaling call analytics workflows using distributed computing platforms like Zerve, facilitating rapid processing of extensive audio data.
Practical insights from using synthetic voice data to develop, test, and validate performance metrics and AI-driven evaluations without privacy risks associated with real-world data.

By integrating advanced speech recognition, powerful LLM summarization, and scalable computing platforms, data science teams can redefine call center analytics, providing comprehensive, timely insights to drive smarter business decisions.

Bio: 

Dr. Greg Michaelson is Co-Founder and Chief Product Officer at Zerve, a young, stealthy startup that’s rethinking the data science development experience. Previously, Greg was an early joiner at DataRobot where he played many roles, including Chief Customer Officer. Prior to that, he worked as a data scientist in the financial sector after earning a PhD in applied statistics from the University of Alabama.

Open Data Science

 

 

 

Open Data Science
One Broadway
Cambridge, MA 02142
info@odsc.com

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Youtube
Consent to display content from - Youtube
Vimeo
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google