SPEECH RECOGNITION   Term Paper ID:32745 Click to get this paper

Essay Subject: Explores the issue of speech recognition including applications forecasts and state of the industry ...... More...

10 Pages / 2250 Words 9 sources, 16 Citations, APA Format $40.00 Return to List of Papers

Paper Abstract: Explores the issue of speech recognition and computers, including applications, forecasts and state of the industry. Discusses speech recognition software. Includes an abstract, table of contents. Paper Introduction: Table of ContentsAbstract Introduction Need for Speech Recognition Types of Speech Recognition Market Penetration Applications Benefits Concerns Conclusion References Speech Recognition Abstract Voice recognition for computers has long been considered the mostnatural way for humans to interact with their electronic tools However bringing about effective speech recognition has proved more difficult thanmany originally thought with the result that speech recognition has yet toreach the widespread implementation once envisioned At this time speechrecognition software is most prevalent in those industries that have a Text of the Paper: The entire text of the paper is shown below. However, the text is somewhat scrambled. We want to give you as much information as we possibly can about our papers and essays, but we cannot give them away for free. In the text below you will find that while disordered, many of the phrases are essentially intact. From this text you will be able to get a solid sense of the writing style, the concepts addressed, and the sources used in the research paper. Speech industry making minor, steady gains. For many years, speech recognition hasheld considerable promise at improving the interface between users andtheir computers. (2 3, September 12). 144.Speech technology. They also speak the region for which they want information,and the system automatically provides the information to them. If the loan is approved, the loan agent closes the loan; ifthe loan is denied, the agent suggests alternatives, such as a lesseramount or a longer payback period. U. In these applications, the computer must not only recognize whatthe speaker is saying, but also verify that the speaker is who he or sheclaims to be. However, speech recognition facesobstacles in terms of speech authentication and other security issues.Continued development is expected in this industry as the cost ofimplementing voice recognition continues to decline, and it is likely thatcontinued improvements will help bring about the widespread implementationof voice recognition that has long been anticipated. Market Penetration According to a 2 2 study, approximately 25 percent of all Fortune 5 companies invested in voice technology in 2 1; this was an increase fromthe 12 percent who invested in such technology in 2 . People are used to using their voice when using the telephone,and the touchtone keypads that have provided access to telephone menus canprove cumbersome and difficult to navigate, particularly as cell phoneshave become more popular. Another concern involves the cultural changes that will be required ifspeech recognition becomes widespread. Banks and other financial institutions implementedprograms by which customers who called the call center are able to speakthe function they want to complete rather than use a touchtone phone tomove through a complex menu hierarchy. As already noted, reducing the number of humanbeings required to staff call centers can result in significant savings.According to some estimates, a typical voice interaction costs a company 2 cents; the same transaction using a human operator can range upward of $1 per call. One of the challenges for speech recognition is the issue of accents.However, as improvements have been made in the ability of computers tomimic various regional accents, similar strides have been made forcomputers to recognize regional accents, as well. According to this reasoning,users will be able to interact with computers in much the same way thatthey interact with other individuals, and the interface would be seamless.However, many speech recognition programs require that users rely onspecific words and commands to accomplish particular tasks, and there aresome who consider that these requirements hamper rather than enhancecreativity, at least as the industry currently stands (Pleasant, 2 2). In addition, the issueof background noise will have to be solved before speech recognition couldbe implemented in this type of environment (Pethokoukis, 2 3). (2 3, February). 2.Thomas, D. Computer Weekly, p. Need for Speech Recognition Speech recognition has been considered a desirable interface optionbetween user and computer since speech is generally considered the mostcommon form of interaction (Pleasant, 2 2). This systemreplaces separate information lines for the company's theaters, and made itpossible for Odeon to close one of its call centers. Similar problems exist with voicerecognition systems used in personal computers, although since thesecomputers are typically located in more controlled environments than cellphones, background noises can often be minimized (Fontaine, 2 2). Types of Speech Recognition There are several different types of speech recognition, each of whichhas different applications in the market. One of the anticipated benefits of speech recognition is thatsophisticated speech recognition will make it unnecessary to learn complexsystem commands in order to use computers. Callers answer a seriesof eight or nine questions; the answers cause a credit check to beperformed and the caller is transferred to a live agent to complete thetransaction. This is a much more difficult process than merelyrecognizing the actual speech. "Hello, computer" he says, and then remarks that the keyboard he has touse is "quaint". CommWeb, n.p.Speech industry. This sample is stored and then used for future contact to ensurethat the voice matches. In addition to providing a more naturalinteraction between customers and the system, telephone voice recognitionprograms also increase the familiarity that users have with such softwareand may help increase the adoption of voice recognition in other types ofapplications in the future ("Speech Technology," 2 2). The opposite ofthis, text to speech, reads text out loud. This system not only saves the company money by havingto operate only one call center, it also eliminates the need for eachindividual theater to record its own tapes each week as films change. ReferencesCastelluccio, M. Our speech is affected by the mood we arein, whether we are sick or tired, and even the various ways that we canpronounce the same word depending on the meaning we are trying tocommunicate. M. (2 2, January). In these applications, voicerecognition replaces touchtone keypad entry. Automatic speech recognition(ASR) allows machines to understand human speech to the point that themachine can transcribe the input from speech into writing. Receiving loud and clear. However, analysts haverecently noted that speech is the most common form of interaction amongpeople, not necessarily between people and machines. One suggestion isthat the Web sites might make only certain languages available for speechrecognition; this is not so different from keyboard input-typically, Websites recognize only certain languages for input, as well (Castelluccio,2 2). Keyboards and mice can be awkward compared with simplyspeaking commands, and some software products are now on the market thatunderstand dictation and simple commands. However, we are a long way fromthe ubiquitous speech recognition envisioned by Star Trek, and some wonderif the promise has not been overextended. Reviving the oral tradition. Introduction When Star Trek IV debuted in theaters, moviegoers were treated to ascene where Engineer Scotty attempts to use a computer in the 2 th century. It offers thesignificant advantage of being hands-free, and can be less cumbersome thantraditional keypad or keyboard entry. (2 3, Winter). Although not yet inwidespread use, voice authentication typically matches a voice against astored sample and determines whether there are enough points of commonalityto verify identification. Individuals who have sought toprogram their cell phones for hands-free use may appreciate the problemsthat arise when the initial recording-against which future requests arecompared-includes significant background noise. Unless the backgroundnoise is eliminated or duplicated, it can be difficult for the cell phoneto recognize the requested phone book entry. Communicate, pp. (2 2, July). Benefits A key benefit to using voice recognition software is the cost savingsrealized by companies. However, even this technology can be fooled bysomeone able to imitate the caller's voice, resulting in a false positive,or when the caller uses different intonation, resulting in a false negative(Eisenberg, 2 3). This is particularly helpfulfor visually impaired users. Individual accents can be overcome with relativeease when compared to the issue of connotations that different dialects ofthe same language possess. Theswitch to a voice-driven system has also proved welcome by customers, whopreviously expressed frustration at using a touch-tone input system to gaininformation (Thomas, 2 2). Voice recognition gives Odeon users 'human' phone contact. Concerns As with other new technologies, there is concern about the form thatspeech recognition will take in the industry. 18.Wagley, J. At this point, companies are using voice recognition to providesolutions for common transactions and requiring callers to either use thetouchtone keypad or interact with a live person for more complextransactions (Wagley, 2 3). (2 3, November 13). The State of speech. While individual applications, such as Adobe Acrobat, couldostensibly learn to recognize and speak in specific languages, the issue ismore complicated when considering how individual users might access aparticular Web site given the global reach of the Web. Applications The financial services industry was an early adopter of speechrecognition solutions. Voiceauthentication is used to verify that a particular voice belongs to aparticular user; decisions are then made to determine which capabilitiesthat user has based on the voice authentication. New York Times, p. Already,integrated cellular phones make it possible for users to speak the name ofthe person they want to call and be automatically connected. Theother popular potential standard is VXML (Pleasant, 2 2). This type of speech recognition has one ofthe highest implementation rates today as many companies have adoptedinteractive voice response to automate their call centers. Is that you, son? includes a text to speechoption that will read documents aloud. The reason that this scene is humorous is because it wasassumed that speech recognition would be the standard interface betweenuser and computer in the future. 22. Table of ContentsAbstract 2Introduction 2Need for Speech Recognition 3Types of Speech Recognition 4Market Penetration 5Applications 5Benefits 8Concerns 9Conclusion 11References 12 Speech Recognition Abstract Voice recognition for computers has long been considered the mostnatural way for humans to interact with their electronic tools. The agent does not have any influenceon the decision of whether the loan is accepted or rejected, only on theaction that is taken subsequent to that decision (Wagley, 2 3). While cell phones and call centers have benefited from voicerecognition, the automobile may be the next highly visible and highlysignificant area where voice recognition makes an impact. Indeed, people areused to interacting with machines without the use of speech, whetherthrough levers-in the case of strictly mechanical applications-or through amouse and keyboard, or touchscreens, in the case of electronic devices.Speech recognition has been accepted by those with certain disabilities whohave difficulty using traditional input devices, however. Information Systems Auditor, p. The greatestlevel of investment in speech technology is in those applications thatprovide customer interface, such as call centers, although analysts expectthat more traditional computer applications will see speech recognitioninvestments in the near future. G8.Fontaine, V. This can result inindividuals re-recording their voice "tags" in order to ensure that thereare understood by the computer. As computers become more integrated into cars, someexpect that drivers will be able to access a variety of different featuresusing voice recognition technology, including weather and traffic reports,tourist information and medical assistance. The system is able tohandle 15, calls per hours and boasts a recognition accuracy rate ofnearly 98 percent. At this time, speechrecognition software is most prevalent in those industries that have a highlevel of direct customer interaction. Institutional Investor, 37, p. Aside from technical standards that may pose a problem to widespreadimplementation of speech recognition, there is also the issue of regionaldifferences in language. Odeon Cinema, based in the United Kingdom, uses voice recognitiontechnology in its call center. Increasingly, voicerecognition is also being used in automobiles, and has gained popularityamong those individuals who find it difficult to use traditional inputdevices such as a mouse or keyboards. (2 2, October 1 ). 59-6 .Eisenberg, A. Many in the industry arecalling for standards that will provide for consistent applicationregardless of which company actually develops a specific speech recognitionprogram, but such standards have yet to be established. When individuals are interested in learningwhat time a movie plays, or how to buy tickets or information, they call asingle phone number and speak the name of the movie in which they areinterested. 13-14.Pethokoukis, J. Key to successful application is the enrollment process. The issue becomes even more complicated whenthe many different languages that exist are considered for speechrecognition. Considerable progress remains to bemade before speech recognition reaches its potential, but companies arerapidly recognizing the advantages of this technology and are investing inits continued development. Microsoft, whichis a late entrant to the speech recognition market, has partnered withSpeechWorks to develop and promote its speech recognition technology, andmaintains that its standard-SALT-should be the industry standard. The simplest way thatcomputers accomplish this is that they learn individual speakers' accents;this works well when there is only one user on one computer, but isdifficult to implement when multiple users access the same computer.Synthesized voices have been taught to speak in regional English accentsvarying from Seattle to BBC English, and some vendors are touting theability of computers to speak in with "authority, clarity and calm" inspecific situations (Castelluccio, 2 2). However,bringing about effective speech recognition has proved more difficult thanmany originally thought, with the result that speech recognition has yet toreach the widespread implementation once envisioned. Interactive voice response istelephone-based technology that allows direct human interaction with adirectory or database system. The Song of the turtle. This 2 2 study also indicated thatinvestment in speech technology is expected to expand to more than $4billion in 2 7 ("Speech Technology," 2 2). This refersto the process of "teaching" a computer program to recognize a particularspeaker. Increasingly,auto makers are incorporating voice recognition technology to control theaudio system, navigation devices and even the suspension settings on high-end automobiles. AI Magazine, 24, p. Strategic Finance, 83, pp. These applications can still encounter problemsif the speaker has a cold or if there is background noise sufficient tomask the voice. America First Credit Union, in Utah, uses voice recognition to enablecallers to complete a simplified loan application. At this point, automatic speech recognition is highly dependent on theconditions of the initial recording. Just as diners and other complainabout the use of cell phones in public places, workplaces have thepotential to become far noisier if speech recognition becomes a commoninterface between user and computer. The final typical voice recognition type is extensiblemarkup language (VXML) that allows users to interact with the Internetusing a voice browser in much the same way that users interact with the Webvia text-based browsers (Castelluccio, 2 2). News & World Report, n.p.Pleasant, B. The possibility of having multiplepeople speaking to their computers-and their computers replying back tothem-could be detrimental to office productivity. This problem may be particularlyacute for those who work in cubicles and who do not have sufficient soundinsulation from one worker to the next. IBM has focused on a technology that requires a speaker toread a selection from a magazine-of the speaker's own choosing-for 3 seconds. (2 2, October). Advocates of this technologycite the advantages of using hands-free access for these features whilecritics charge that the driver's attention-if not their hands-will still bedistracted (Fontaine, 2 2). The End of keyboards. More than three-quarters ofthat growth is expected to be in call centers and business portals, whichhave seen the most rapid acceptance of this technology in the past, as well("Speech Industry," 2 3). S. Currently,investment in speech recognition and other types of speech technologyremains low relative to other types of information technology investment,but the return on investment is considered relatively high. Adobe Acrobat 6. Conclusion Speech recognition is not necessary a natural way for humans tointeract with machines, but it can prove convenient. It is not surprising that telephoneportals have proved a successful entry point for speech recognitiontechnology. (2 2, November 26). This research considers speechrecognition, including its application and current state of the art as wellas its future direction. If this paper is not what you are looking for, you can search again: Search for: or Click here to request an essay written just for you.

Many of our Papers can be Downloaded From This Site!

PLEASE READ THIS, IT IS IMPORTANT!

Office hours are Monday through Friday, from 9 am to 5 pm (PST). You may place orders for custom research over the phone during office hours. E-mail requests can be made to our graduate and undergraduate department any time, and will be reviewed during office hours. You may also contact customer service any time through e-mail, and we will review your message during business hours. A great many papers can be downloaded right from this site, but not all of them. If you would like to know if a particular paper is downloadable, just look in the description for: "Available for Internet Download: Y" or "Available for Internet Download: N" If you wish to purchase a paper which is NOT available for immediate download, you will need to make other shipping arrangements. Also, please be aware that these orders are processed Monday through Friday from 9 am to 5 pm (PST). If you place your order after 4:45pm on Friday, it will not be processed until the following Monday morning. We charge $8 per page for all of our pre-written reports, plus shipping (and tax for California residents). However, the highest cost of any ONE report is $136, or 17 pages. Please, take a moment. Make sure you have chosen the report you want or need BEFORE you complete your order. If you are not sure, allow us to help you. We do not offer refunds or exchanges, so it is important for you to let us answer your questions during office hours. Reports which are e-mailed or downloaded are in Microsoft Word format. We are making more reports available for e-mail delivery faster than we can update our listings. Please call to check on the status of particular reports. There are many other shipping options which are listed on the Checkout page.