TrendMiner 5.0 includes a bevy of rich functions that help call center managers search and mine large audio archives of agent-customer interactions to glean out valuable business intelligence, trends and insights.
New features bundled into TrendMiner 5.0 include: a search refinement tool, wizards to assist in the definition of call filter attributes and word matching algorithms, a revamped call payback and snippet engine, new charting and trend reporting tools, and richer summary statistics to help administer large call databases.
The new release also beefs up performance and accuracy which industry analysts agree are critical to successful large-scale deployments of speech analytics systems.
Officials at Fort Myers, Florida-based company boast that TrendMiner can turn audio files into mineable information at 16 times real-time and retrieve a phrase from call databases at break-neck speeds of over 11 million times real-time playback.
Officials also argue that TrendMiner is more accurate than conventional phonetic-only technologies since its patented speech recognition engine combines patented speech algorithms, natural language processing, statistical methods and advanced pattern matching.
CallMiner claims its approach allows over 70% of calls to be transcribed accurately enough for analysis; which is at the high-end of the accuracy scale today.
TrendMiner is part of CallMiner’s modular Analytics Suite which also includes two other components: a Core Engine (speech indexing and classification) and Tireless Supervisor (automates call scoring).
