IVR Deconstructed

We live in an era that is saturated by constant and continuous news updates, whether you want the news or not.  The amount of financially themed news stories has increased exponentially in the last three years due to the precarious state of the economy.  Just this week, there has been a variety of fiscal stories that have made front-page news.

The debt crisis is looming, with fierce debate ongoing concerning the debt ceiling and whether it should be increased, and if so, to what amount.  The amount of permissible public debt is set at $14.3 trillion dollars as stipulated by federal law.

That debt ceiling was reached on May 16th, at which point special measures were undertaken to extend the government’s ability to make payments until August 2nd, at which point the U.S. Treasury would no longer be fiscally solvent.  This would mean that the U.S. Treasury could not borrow additional money to pay bills, and while capital could come from alternative sources, the government would be forced to choose which obligations were of the utmost importance.

The President and Congress are in a stalemate over whether to raise this debt ceiling, with House Republicans refusing to raise the amount without some sort of deficit reduction and the President and Senate Democrats refusing to acquiesce to these demands.

Pictures surfaced yesterday of what many are hoping is the iPhone 5, which is supposed to be released sometime in September.  As anyone who even casually follows the tech industry knows, the release of a new and improved model of the iPhone results in widespread excitement and a massive flurry of people attempting to get their hands on one.

This of course means that Apple’s shares, which are already closing at their highest point ever ($403.41/ share), will skyrocket even further.  Any savvy financial investor wants to be informed immediately of activity that can potentially affect their portfolio.

While many smartphones come equipped with stock monitoring capabilities, many people continue to have feature phones.  Also, checking on stock quotes, currency conversions, portfolio tracking and monitoring financial news can be tedious and often times dangerous.  That is where an interactive voice response system can come in handy.

An IVR system can be easily accessed via any kind of phone (mobile, landline) and can be used hands-free as well.  Users can simply call in, speak or type their account information, and have all the financial information they require immediately at their fingertips (or more accurately, eardrums).  IVR enables users to stay continuously updated on their financial portfolio while on-the-go, without dangerously trying to read and commute.

Mobile phones are a vital part of how we function as a society today.  According to mobile statistics collected by Mobithinking.com, there are 5.3 billion mobile subscribers in the world (77% of the world’s population).  The sale of mobile devices steadily rose in 2010, with smart phones showing the strongest growth.

Eight trillion text messages will be sent in 2011.  Over 300,000 mobile applications have been developed in the last three years. Mobile Applications have been downloaded 10.9 billion times.  Mobile phones are used to surf the web, run applications, email, text, play games, bank, shop, receive news updates, watch videos, participate in social media, listen to music, and of course, make phone calls.

Based on the above data, it’s clear that users rely on their mobile phones, whether feature or smart, to perform an enormous number of tasks for them.  Mobile phones have become one of the most vital communications devices for both personal and professional use, and all signs point to mobile phone usage continuing to soar.

While this mobility is infinitely attractive to mobile users, there might be an extreme downside to this freedom of technology.  At the end of May, the World Health Organization (WHO) released a report that classified mobile phones as presenting a possible risk for a specific type of cancer in humans.

According to a joint report released by the WHO and the International Agency for Research on Cancer, mobile phones show limited evidence of carcinogenity.  The WHO has said that:  “A positive association has been observed between exposure to the agent and cancer for which a causal interpretation is considered by the Working Group to be credible.”  Scientists have theorized that this could potentially lead to a malignant type of brain cancer.

However, a new study released today by the Journal of the National Cancer Institute detailed how there was no link found between mobile phone use and cancer: “The absence of an exposure-response relationship either in terms of the amount of mobile phone use or by localization of the brain tumor argues against a causal association.”

So which is it?  Is there a heightened risk of developing a brain tumor from extended cell phone use or is there no risk at all?  These conflicting reports may leave mobile phone users confused.  Should they limit their talking time?  Should they feel free to gab away?

Only a slew of additional research studies will be able to concretely prove the effects of prolonged cell phone use one way or the other (remember the cell phone only came into popular usage in the late nineties).  In the meantime, there’s a plethora of alternative uses for mobile phones, and users can email and text in order to stay in constant contact while simultaneously minimizing the risk.

Yesterday’s blog discussed the developmental history of natural language processing (NLP) technology.  From the 1950s to today there have been various developments in the field of NLP that have made it a critical component of speech recognition technology and have enabled application development for myriad devices.

Speech recognition is only one of the functionalities that use NLP.  Other crucial tasks that NLP performs include automatic summarization (production of a readable summary of a text or a portion of a text), machine translation (the automatic translation of text from one language to another) and question answering (the capability of answering human-language questions).

Speech recognition is considered the opposite of text-to-speech, which is the process by which a computer system converts typed text into spoken words.  Speech recognition is the conversion of spoken words.  Voice recognition is featured in many interactive voice response applications and is helpful in enabling hands-free phone usage.

Often times an automatic speech recognition (ASR) functions more smoothly when the recognition system has been trained to detect the voice of a particular speaker.  However, speech recognition also works for multiple speakers, as the application is designed and programmed to recognize speech in general, without being customized to a single speaker.

There are many ways in which ASP can be effectively used.  It can be programmed into car or home audio systems in order to enable hands-free communication (voice dialing), appliance control (both issuing and processing verbal commands) and even data entry (users can submit credit card, billing or shipping information by simply reciting their info).

One practical use for NLP as documented by the Belgian technology consulting firm Nmahn is to conduct searches of both the Internet and computer databases.  Many search engines rely on Boolean searches to increase their searches.  Boolean searches are typically defined as ones that have tow data values that are true or false.

However, most search engines today are powered by NLP as opposed to Boolean, giving users what is traditionally thought of as a more friendly search engine experience.

Natural language processing (also referred to as NLP) is the study and development of interactions between computers and human languages.  Most speech recognition technology utilized in interactive voice response (IVR) applications is based on natural language processing research and technology.

While NLP began as a facet of artificial intelligence, it has escalated into the exploration of how humans and computers interact. As anyone who has utilized speech recognition on a computer knows, it is often times difficult to speak in natural language (that is unaffected or unparsed) and have a machine be able to successfully recognize it.

A quick example: There are dictation programs that allow people utilizing a microphone to speak their words instead of typing them.  This is especially important to those who might not have the physical capability of typing for one reason or another.

Those who are unable to type rely on microphones that have the ability to interface with a computer and enables them to dictate their documents.  However, this process can be quite burdensome because a computer does not have the innate ear for language that a human does.  For example, instead of hearing “and,” the computer will hear “an.”  Instead of interpreting the proper name “Jill,” the computer will incorrectly type the word “gel.”  Obvious problems arise from the misinterpretation of natural language.

Natural language processing is a field that focuses on the linguistics of computation with the intent of improving and streamlining human-computer communication.  NLP is generally recognized to have originated in the 1950s, when scientists developed methods to test a computer’s ability to impersonate a human in real time.

The field has continued to develop, with scientists integrating knowledge from various disciplines (linguistics, computer science, statistics, et cetera.) in order to refine and improve machine translation, text-to-speech and language technology in general.

In our continuing analysis of unified and telecommunications, we will be examining the field of NLP and how it significantly improves the way in which human beings interact with machines.

There are so many things to talk about today I thought I’d just do a hodgepodge post to cover them all.

First a couple of humanity items…

The Norway thing. I actually don’t even have words for that, so I’m not going to address it other than to say what a senseless waste, which I’m sure is what most people are thinking.

Also, singer-songwriter Amy Winehouse died over the weekend. They don’t know the cause of death yet. Regardless of how she died, it’s a shame to lose a young artist. I for one don’t think there are enough artists in the world.

On the telecom front…

RIM, makers of BlackBerry smartphones, is planning to cut 2,000 jobs as the company continues to lose market share to Android and Apple.

According to Bloomberg, RIM is cutting one-tenth of its workforce (down to 17,000 total employees) to “focus on areas that offer the highest growth opportunities.”

I’m not exactly sure what that means, but it’s not the biggest surprise to see the company cut back. Apple and Android are taking over the market. BlackBerries are even starting to lose out to Androids and iPhones in the business sector, BlackBerry’s bread and butter.

On the domestic and world economy…

U.S. lawmakers have yet to raise the federal debt limit, keeping the nation at risk of defaulting. In response, the stock markets have dipped a bit. Nothing to be alarmed about, but illustrating some uncertainty nonetheless.

To make matters worse, Greece seems likely to default on its bailout loans despite austerity measures (and subsequent protests and riots). The European Union is planning a debt swap to ease Greece’s burden—adding as much as 30 years to the term, according to the BBC.

But rating agency Moody’s has dropped Greece’s rating to just above default. “The announced EU programme…implies that the probability of a distressed exchange, and hence a default, on Greek government bonds is virtually 100%,” the agency said.

Unfortunately, some believe a Greek default could affect the U.S. economy, including former Fed chairman Alan Greenspan. He said it could send us into another recession.

For all our sakes, I really hope he’s wrong.