Speech Analytics: Completing the loop of data mining

Infotech Lead Asia: Amit Bishnoi, VP – Applications Business, AGC Networks, says the Indian Contact Center is one of the true successes of the India growth story and has played a significant role in putting the country on the global map.

Furthermore, the industry has constantly evolved to become a valued partner and remained strong and exciting even during the recent economic downturn. Technology has played a major role in this transformation by driving efficiency in operations, reducing cost and streamlining process to achieve excellence.

(source: planet-pott.de)

With the ever increasing volume of audio calls in contact centers, it is of paramount importance to maintain quality and also gain valuable vital customer insights. This intelligence can be further leveraged to not only drive customer satisfaction, but also re-align business strategy, process andproducts to best suit the customer’s changing requirements. Speech Analytics, also known as audio mining is one software technology that converts unstructured conversations into structured output. This structured information helps organizations to truly understand what the customer needs and helps to proactively react.

Most of the speech analytics solutions take their input from recorded conversations. Recordings can come from any vendor that is able to capture and share them; this includes speech analytics vendors, stand-alone recording vendors, workforce optimization providers, unified communications vendors, hosted contact center infrastructure vendors, etc.  The primary output from these solutions is metadata (files of data) in two general forms: phonetic representations of a conversation, or a transcript of a conversation.

Amit Bishnoi, VP – Applications Business, AGC Networks

Benefits of Speech Analytics Solutions:

Root cause analysis

Understanding the reasons why people call. This will shed lights calling patterns, questions asked during the call, etc. This will enable deeper insights into customer’s interest level in particular product or service.

Trend analysis

Identifying anticipated and unexpected reasons why customers call. This can additionally help in training managers to get equipped with firsthand information while imparting training to the team.

Emotion detection

Understanding callers’ emotional state (some solutions can measure emotional changes and heightened states).

Talk analysis

Understanding caller and agents talk periods, including measuring silences, holds, transfers, talking over, etc. This can add significant value by identifying call handling abilities of executives, knowledge of executives about a particular question

Script adherence

Monitoring how well agents follow their scripts, communicate required information, or say inappropriate phrases or words.

Quality assurance

Identifying conversations that require management attention. As this solution provides real time analysis of conversations, the management is better abreast with the cause of the problem. This will help in fastening up the overall decision making process. Additionally, this will also provide useful information an organization is perceived. Speech Analytics will certainly help in improving the products and services delivered by an organization.


This will enhance the understanding in how well staff members perform their jobs with each call they attend. Additionally, it can be decided if additional training is needed.

Additionally, Real-time speech or interaction analytics applications that address both speech and written interactions (chat, email, short message service (SMS)) are expected to play an increasingly important role in contact centers. Real-time speech analytics solutions can also be used to identify situations where agents do not appear to be in compliance with proper procedures, and provide them with information to help them rectify the situation while the customer is still on the line. For example, if an agent is supposed to verify a caller by asking specific questions but skips this step, a real-time speech analytics application can pop up a screen or send an alert to remind the agent to do the verification.

Market Trends:

Speech analytics solutions have performed extremely well over the past 18 months, withstanding the pressures of a recessionary economy. The positive trend for speech analytics is expected to continue in 2013 with North America remaining dominant user of speech analytics, while UK and other parts of Western Europe and India market expected to gain momentum. Furthermore, a growing number of end users who piloted speech analytics are expanding their use of the solution by adding seats and rolling it out to other departments within their company. This is reminiscent of the fact that apart from contact center, enterprise departments adopting speech analytics include operations, sales, marketing, product development, risk management and fraud detection. Additionally, Collection agencies are adopting speech analytics to measure adherence and to identify collections best practices

Product Trends:

WFO vendors continue to integrate speech analytics modules into their suites. At the same time, the best-of-breed speech analytics vendors are adding WFO functionality to their offerings. While this is an important trend, the most notable product-related development, which will accelerate over the next several years, is the evolution of speech analytics into multi-channel analytics platforms, as information from text-based sources (emails, chats, online forums, etc.) is increasingly incorporated into the analytics mix. Also witnessed is that stand-alone speech analytics vendors are increasingly providing their own recording capability and, in some cases, are moving towards delivering a WFO suite.

As India is logging on high growth in contact sectors market, Speech Analytics Technology will provide significant advantages, enabling real time access to important data buried audio format. Speech Analytics provide huge opportunities for organizations to understand the relationship between organizations and their customer and prospects. No matter, Speech Analytics is in its adolescence stage, but once applied it will have significant impact in applying decisions, supported with real time data buried in conversations.

Amit Bishnoi, VP – Applications Business, AGC Networks
[email protected]