Introduction
“Speech recognition is the ability of a machine or program to identify words and phrases in spoken language and convert them to a machine-readable format” ("Speech recognition ," 1997). In today’s world, automatic speech recognition has been taken so lightly that people do not notice it how great technology is or how often we use it. Being that its creation was only a few decades ago and that it has come this far, there is still so much room for advancement in years to come. Today we use this software for many things such as call centers when we call companies and corporations that receive a lot of incoming business or with the blind or authors when they speak into a microphone and what they speak is made into a document. As we take a deeper look into the creation of automatic speech recognition we will address its possible disadvantages and advantages that will give us a better understanding of this new technology.
Background Information
As stated before, automatic technology was only launched decades ago, to be specific, in the 1950s. “Bell laboratories designed in 1952 the “Aubrey” system, which recognized digits spoken by a single voice” (Pinola, 2011). Notice that only a single voice could be used once when first tested on the recognition software because technology was not advanced enough to understand more than one pitch or even different sexes. Numbers are also the most general form of communication when speaking verbally because they only represent quantities rather then expressions and feelings. Numbers can be communicated in many other forms even with quick universal hand signals of holding up your fingers. Not until 1962, did IBM create its “Shoebox machine which could understand sixteen words spoken in English” (Pinola, 2011). Transitioning from numbers to words was a huge advancement because those sixteen words were the first step in the human language with basic commands, which in the spectrum of an answering service is a huge help because it can decide whom the caller needs to talk to. A major step was in 1976 when the “U.S. Department of Defense funded the DoD’s DARPA Speech Understanding Research (SUR) program…it was responsible for Carnegie Mellon’s “Harpy” speech understanding system. Harpy could understand 1011, words, approximately the vocabulary of an average of a three-year old” (Pinola, 2011). At this point in time, a development on interpretation different voices was made as well as understanding fully formed sentences. The Department of Defense funded this research and program in hope for their benefits to use for themselves. Their funding was obviously beneficial and brought much progress to past research. Years went by and automatic speech recognition worked its way further and further into commercial applications such as dictionaries and the business world until today its part of our everyday activities.
Potential Benefits
Automatic speech recognition has brought organization to problems, efficiency to the business world as well as super power like abilities to people in need. With its range of skills, it can be used for, a variety of business and medically progressions can expand upon it. As stated by Source Security, “voice recognition is non-contact, non-intrusive and easy to use” (Schelps). This software allows a representative to deal with multiple customers at once, but also be hundreds of miles away but still get the same information or even purchase products bringing in income for the company while using automatic speech recognition decreases expenses because the company would need less on call representatives. Instead companies can use that use those representatives to deal with more in depth problems and questions and use the software for simple task and issues that can be handled with simple statements and directions. Also the software is very direct and simple to use so people of all ages can use it and benefit from it without needing instructions, its an automatic understanding. The elderly, who are not technology updated, do not need someone to teach the terms and the process of automatic speech recognition, it an automatic understanding. For the blind disabled, they are able to speak into a microphone and have their words spoken appear on a computer possibly in a document if they want to write a book or a paper for school with out the need for assistance. Where as before they would need to speak their thoughts as someone else typed them on a computer for them. Also with commands they can tell the computer to do things with out having to see the computer its self and choose with their hands and eyes but rather with their vice. For security benefits, the ability that automatic speech recognition can understand both letters and numbers allows verbal passwords to be created with mixtures of the alphabet and numbers on top of the security system recognizing the voice, which allows them to be of a much higher level of security than with just entering in digits and numbers. Hackers are always becoming smarter and smarter and we need technology and security to keep up with them and be a step ahead of them in order to keep our selves protected.
Potential Problems
All new advancements come with their flaws and in the case of automatic speech recognition, there is quite a few. First of all everyone’s voice are different, that is what makes us unique but it also generates problems. “The fundamental frequencies of speech sounds uttered by woman are about an octave higher than those of men, and those of children are even higher” (Ainsworth, 1939). Not only does voice differ person to person but it also depends on our age and our sex. When a toddler, they are just learning to speak so they do not pronounce sounds and words properly, and then children hit puberty where if a boy, their voice will squeak and change dramatically over a short period of time until settled into their so called set voice for life. How is voice recognition suppose to understand all these different stages of growth all at one time on top of accents and mumbling, it cannot. Also, with so much migration there are so many different accents; "People from different parts of the country and different social an economic phenome from another in certain words, and changes in the rhythm and inonation of the utterance. This variability is even greater with people speaking a second language. The competence with which it is spoken depends on the motivation, intelligence, and perceptual and motor skills of the speaker, and also of the age at which the second language was learned (Ainsworth, 1939)." With the example that Bostonians do not pronounce their “R’s” and that New Yorker’s replace their “E’s” for “A’s,” its merely impossible for technology to comprehend accents of Americans on top of other countries if real live people have a hard time understanding each other. In addition, white noise is an issue, “background noise as the caller interacts with the system,” may give false triggers to the recognition or the software may not recognize at all what the speaker is saying (Schelps). There are still many flaws to be adjusted to the program to make it a more credible piece of software.
Conclusion
Technology is constantly advancing day to day and the more it advances, the more humans deal with technology rather than with humans themselves. Technology is able to deal with minor tasks and problems allowing humans to deal and spend more quality time on bigger, more important issues. Automatic speech recognition allows a cheap, efficient way for businesses and commercial products to communicate with their customers without spending money on representatives on harmless information like business hours. Any new technology will have flaws but automatic speech recognition is definitely a positive advancement and will continue to benefit the task of multitasking. Flaws will continue to decrease and improvements will continue to increase just like in any new in invention.demonstration of automatic speech recognition
Reference Page
Ainsworth, W. A. (1939). Speech Recognition by Machine. (p. 51). London United Kingdom: Peter Peregrinus Ltd.
This Source is Valid because the booked is dedicated entirely to the process of automatic speech recognition and how it is made from the noises that come from our mouth to the responses the computer makes when it hears the noises. The book addresses pro and cons of speech recognition which I also talk about in my research paper.
Pinola, M. (2011, November 02). Speech recognition through the decades: How we ended up with siri. PCWorld, Retrieved from http://www.pcworld.com/article/243060/speech_recognition_through_the_decades_how_ we_ended_up_with_siri.html
February 23, 2012
PCWorld is a online magazine which specializes in technology, they actually sell products themselves so they obviously know what they are talking about. This magazine discusses other similar topics such as tablets, phones, and laptops. The website has privacy policy and “contact us” section to reach them for questions and concerns.
Schelps, D. (n.d.). Voice recognition – benefits and challenges of this biometric application for access control. Retrieved from http://www.sourcesecurity.com/news/articles/co-3108- ga.4100.html
February 23, 2012
Source Security talks specifically about security details and the security market. They sell different types of security packages and discuss why their security is great which is where they discuss automatic speech recognition and it advantages and what it does. They are a well known and established industry.
Speech recognition . (1997, June). Retrieved from
February 29, 2012
Tech Target is a valid source because is discusses my particular website down to the definition and it also gives tutorials about other technological problems. It has expert advice as well as blogs for outside opinions and advice. This website gives direct definitions for terms used for my topic and for research paper. These are valid because by the definitions they shows, “people are read” articles to relate similar topics.
Speech interfaces are ready to listen. (2001, October 22). CNN. Retrieved from http://articles.cnn.com/2001-10-22/tech/speech.interfaces.idg_1_speech-recognition- speech-interfaces-phonemes?_s=PM:TECH
February 24, 2012
CNN is a credible source because they are a well know news source. They site their sources when they use outside information, and it is known around the world making it an international credible source. CNN is constantly updating their stories and following new topics with the media.
