Voice Recognition Software - Research Proposal Example

VOICE RECOGNITION SOFTWARE   Voice Recognition Software Introduction Voice recognition is a form of datainput into a system such as a computer using voice. It is an alternative to using of other forms of inputting data such as the use of keyboards, touchpads or even mouse and jokey sticks (Clark, 2002, p.15). In this research, the main topic is based on voice recognition software and how it can be used as a better and choice in real life application. Problem In the security sector, it is important to have a system that is leak prove. The access by unauthorized personnel possesses great threat to the integrity of the organization. Facilities such as financial institutions in particular have strict rules on personnel clearance. The vaults and financial files and data are only accessible by authorized people. They have, therefore, set up measures to ensure that this security protocol is observed which include the use of security keys and feeding in passwords before accessing certain areas. Related Work and Experience The problem above is not a new finding, but an issue that is in every financial institution. This has seen innovators and designers come up with solutions that have not solved the problem substantially. There have been inventions such as the use of security cards that are assigned to the various personnel (Clark, 2002, p. 26). These cards are as special keys, where each has a feature that is intended to enable access to certain places to the personnel concerned. They only need someone to swipe the card in a barcode reader at the door and input a password before a door opens. However, the problem with security cards is that they can be stolen or get lost. Criminal or personnel gets his/ her hands on the card and has the knowledge of the password; a security breach can occur, and it may lead to lose of crucial data or even money.Therefore to reduce the risk of lapses in cases where these security cards end up in the wrong hands, putting that power in somebody’s voice is the only sure way. The voice of a person activating the opening and closing of a door can go a long way in preventing such breaches. This is because voice recognition does not require any password to be fed to the system and it a persons voice is unique to themselves. Background Voice recognition is a technology that involves the translation of spoken words into texts. It is an invention that has transformed the way people interacts with technology from simple gadgets like the smart phone to complicated sets as voice controlled military missiles and fighter planes (Clark, 2002, p.59). Voice recognition has offered solutions to many real life issues ensuring the need to feed a password into the system is eliminated. It is faster and efficient thus common to find voice recognition applications in nuclear facilities, financial storage facilities like safes and space mission centers as NASA facilities. With improving technological advances, such services can be available to the ordinary person to improve his/ her way of doing things. Solution Voice recognition software involves an array of code whose work is to recognize the voice. It has a set sequence in which it executes its mandate and involves the steps below (Clark, 2002, p.103). First it picks up the voice of a person and analyses the sound. This, however, requires a pre-installation training on the part of the users such that the authorized personnel read a phrase provided by the software. This helps the software categorize ones punctuation and vocal ability and save that data in its database. This happens because individuals have a different way on how they pronounce and sound off words. The next step is where the software picks the spoken sound and compares it to a predefined database of words. The software database has a wide range of words that act as its dictionary. If the word matches a certain word, it is displayed as feedback so that the user can be assured the system understands what he/she meant. However, if the word is not in the database, the one nearest sound to the spoken sound is displayed. The last step involves the system using the data now to initiate the required task the required task. In this case, the recognition of the said word such as open and close initiate a motion of the door to open and close respectively. This stage is very important since it is what affirms that the voice recognition software has worked. Evaluation Every system has flaws in one way, or another and the fastest way to identify these offers a chance on improving them. Thus, the voice recognition software has to be evaluated to ensure it is efficient (Clark, 2002, p.370). It involves checking for issues that may arise in the case of unrecognized voice. This is a realistic assessment as there might be the presence of background noise or a person having a sore throat. Can this affect the working of the software? The effect of background noise on the software should be assessed and measured. If the contribution of background noise corrupts the normal working of the software, proper measures should be set up. This can vary from minimizing the noise to using high definition microphone systems that can filter out these noises. The other issue to be assessed should be the capacity of the database (Clark, 2002, p.311). Since it can update in cases where new persons use the system or new words are used, storage space is important. Availability of measures should be in place to ensure there is no data redundancy and that this does not cripple and destroy the system. The next important evaluation involves analyzing the speed of the system and how well it responds to the inputs (Clark, 2002, p. 319). If the software is slow, it calls for improvement as it might be an indication of future system trouble. This should, however, be differentiated from impaired speech on the user side and such instances should not be used to rule out its efficiency. Experiments The experiments on the voice recognition software would primary involve using it together with other technology to improve and build on its usefulness (Clark, 2002, p. 323). Combining voice recognition software together with a motion detector can assure the system as a whole is not compromised. The unit should be set up such that when personnel approach the door, the voice recognition software becomes initiated and wait for the next instruction from a person. This would ensure, energy efficiency since the software would always be switched off until the time a person approaches. Bibliography Clark, L. (2002). Voice Recognition with Software Applications. New York: McGraw-Hill. Read More