Case Study: CallMiner boosts transcription and scorecard accuracy with Macrosoft

Capturing Call Fidelity Impact on Transcription and Scorecard Accuracy

The customer, CallMiner, faced the challenge of low call recording fidelity negatively impacting the accuracy of its automated speech analytics platform. Poor quality audio led to problematic transcriptions, which in turn reduced the effectiveness of the platform's features and scorecarding capabilities, preventing the customer from unlocking valuable business insights from its customer interactions.

The vendor, Macrosoft, conducted an analysis to quantify the impact of fidelity on transcription accuracy. By comparing low and high fidelity call samples, Macrosoft demonstrated that improving audio quality from 24kbs mp3 to 128kbs mp3 resulted in a dramatic 41% increase in transcription accuracy, a 6% reduction in false positives, and a significant decrease in the need for error-prone aliases. This proved that enhancing call fidelity is a critical first step to achieving maximum value from the CallMiner platform.

CallMiner