Using Hadoop to power speech analytics

Business interest in using speech analytics is growing, as contact centers, pathology researchers and others look for more ways to leverage Hadoop big data into actionable insights. With Hadoop, processing unstructured data in the form of voice recordings becomes possible, allowing organizations to track information in phone calls and create solutions to long-standing problems.

One of the areas that could especially benefit from this technology is the corporate call center, according to a recent Enterprise Apps Today article. Businesses have recognized the value of being able to gather and analyze information coming from customer interactions in the call center. Using speech analytics tools, executives can track caller opinions and use this information to improve service quality, increase customer loyalty and ultimately drive revenue.

"Tracking the customer journey and understanding their experience has an effect on everyone from senior level executives to marketing to product development and being able to harness the 'voice of the customer' [VoC] has impact far beyond just the call center," speech analytics professional Jon Ezrine told the publication.

Speech analytics tools that leverage big data can help businesses identify the type of interactions that lead to customer frustration, Enterprise Apps Today noted.

Hadoop and the Parkinson's Voice Initiative
The same tools might also be valuable for driving initiatives such as the Parkinson's Voice Initiative. The project uses an algorithm that analyzes voice tremors to diagnose potential Parkinson's patients. According to The Verge, the project currently has an 86 percent accuracy rate when used in blind testing, and it is 99 percent accurate in the mid to late stages of the disease.

By building a database of voice samples, researchers can improve the algorithm. In a recent interview with SQL Server Pro, Microsoft data expert David Campbell noted that the initiative is a great example of how Hadoop could synthesize a MapReduce job by pulling data output from voice files. As interest in the applications of speech analytics grows, organizations may find value in undertaking such projects of their own.

Categorized by :

Leave a Reply

Your email address will not be published. Required fields are marked *

Hortonworks Data Platform
The Hortonworks Data Platform is a 100% open source distribution of Apache Hadoop that is truly enterprise grade having been built, tested and hardened with enterprise rigor.
Get started with Sandbox
Hortonworks Sandbox is a self-contained virtual machine with Apache Hadoop pre-configured alongside a set of hands-on, step-by-step Hadoop tutorials.
Modern Data Architecture
Tackle the challenges of big data. Hadoop integrates with existing EDW, RDBMS and MPP systems to deliver lower cost, higher capacity infrastructure.