The Effect of Automatic Speech Recognition Systems on Speaking Workload and Task Efficiency - Assignment Example

Comments on ‘The effect of automatic speech recognition systems on speaking workload and task efficiency.’” Abstract: With the development of technology, has come many advantages that have advanced the human thought and research through huge leaps. However, the basic tools of technology may have left a trail of health issues and other associated dysfunctions that may have lead to prominent diseases and disabilities in the longer period of time. The ASR software and other speech recognition systems that make the users tone an unnatural speech pattern may be associated to such long-term affects on the human system. The paper relates such a possibility and examines research that may be relevant to the affects of prolonged use of such tools. Introduction: The research was focused upon the learning of the stigmas that were involved in the processing of speech in automatic speech recognition (ASR) software. The software was used with the subjects that had spinal cord injuries. These results were compared to other subjects that did not have spinal cord injury to evaluate the possible working of the ASR software. The study was conducted to research the possibility of the increase in exertion of functions in the laryngeal system with the prolonged use of the dictation with the use of the ASR system. These functions have been noted by many other researchers and have been cited in Vincenzi et al (2004), Huey et al (1993) and Robertson (1996). The use of the ASR system was also found to be more effective in terms of the number of words that were reproduced as compared to the same words that were tried without using an ASR system. It was found that the ASR system was a more efficient addition in the methodology and that without its incorporation, the system was found to be far less efficient and required a more tiring procedure. The ASR was targeted to be used to substitute the mouse and keyboard attached to a computer and was made to function to perform tasks that requires inputs from either of the devices. The ASR system was tried as a replacement to other conventional input devices and the relevance was established as a comparison over cost of the ASR system- as it was close to what a mouse and a keyboard would be priced at. The system was studied to be of use and assistance to people who have disabilities in terms of a high-level spinal cord injury or a repetitive strain injury at the upper limbs. These disabilities would make the use or the extended use of a normal computer very difficult to these people and such a use may deteriorate their condition even more. Furthermore, as a secondary objective of the research to was also tailored to be seen as a substitute for all the secretarial dictation and work that is required in offices to explore the possibility of the ASR system being a replacement for these additional costs that the organisations have to incur (ES, 1989). In either of the two objects, the study has been related to the impact of using such systems on the health of the voice of people who intend to use this utility. The impact on the voice has been discussed and narrated to the changing of the way the user adapts to the system which has been found as not a normal and natural way in terms of as one person would talk to another (Britton & Glynn, 1989 & Reid et al, 2005). This may have had an effect on the natural way the respiratory, laryngeal and articulatory systems and eventually may lead to a vocal disorder. Method: The research was conducted to be based on two main groups of people that were spread over age. The subjects were divided into those who were handicapped with specified health disorders and those who had no disabilities at all. A total of twelve subjects were studied form which half of them had no health issues and the other half had been diagnosed with and still had a cervical spinal cord injury dating back to at least a year before the study was done. The study was conducted to match the age, height and other physical attributes of the members of the two groups to match in relation to each other so that there was no variance in the condition of the internal human system that was going to provide substance for the research being conducted. All the subjects were tested to have normal characteristics of speech and voice and were reported to not to have any respiratory infections. Moreover the subjects were authenticated as non-smokers and had perfect vision. The subjects were asked to use standard ASR software along with two microphones- one would be used to dictate and the other for acoustic analysis of the individual subject when they were using the software. The subjects were also monitored through medical equipment to measure the changes and any other transformations relating to their human functions and system. The results of the analysis using the ASR software and then without the ASR software concluded that the subjects had the capacity and actually were able to say more syllables in the specified time without the use of the ASR software. This difference was of over one hundred and fifty parent concluding that the use of ASR slowed down the speech of the subjects considerably. The time span considered for this conclusion was one breath group to relate consistently to the other specifics of the research. Comparison over the software tools in terms of the number of breath groups that the subjects had to take were also studied in the conditions of dictation. It was observed that the subjects, when they used the discrete-word software, used a much higher value of breath groups than using ASR. It was furthermore concluded that the breath groups decreased even further when the subjects did not use any tool at all and spoke in natural conditions. This showed that the subjects breathed more and in an unnatural way while using a speech recognition tool and would be harmful for health. Similarly the results also stated that the subjects suffered increased levels of apnea as they used ASR or discrete-word software. This implied that the use of any tool that took dictation would affect the user adversely in the long run and may develop complications in health. This was also added on by the extended time needed for the ASR software in comparison to others as more stress was notices in the functioning of the human system. Research Plan: The research may be extended more in relation to the effect if speech recognition tools on the human system. As the research had not been that conclusive in terms of valuable results it may be relevant to expand the sample size- which was studied initially. The expansion in the number of subjects that can be further studied may provide a more comprehensive data to be analysed and a precise conclusive analysis may be made to relate the factors of stress in terms of number of breaths in the speech patterns (Karwowski, 2006). Along with a bigger sample size, it may also be important to include a broader range of subject sin the study to narrate different ages with behaviour and other social norms. This extended sample in terms of diversification may also give a data range that may be assessed in terms of how human speech systems may be narrated to social background and the relevant sub cultural activities. These results may be more applicable in understanding the cultural and social influences in order to determine the affects of any Speech recognition tool and its use over short or extended time (Karwowski, 2006). Moreover, the research may also be signified by testing and collecting the data in multiple environments- which would be more work and task specific. There may be instances where the work environment may influence in wither increasing the stress levels on the human speech function or decreasing the stress level die to the environment and the facilities. The perception of the environment in the brain is also imperative to know as many influences in terms of stress and other body functions relate to the type of environment (PHFDTDSIS, 1997). In terms of environment, the study may also be extended into a more practical application of the specific environments for which the studies were conducted in terms of the application of the software. The study of the breath group and its analysis maybe clearer in a contact environment where just the stress of the research itself may have diluted or affected adversely some of the data collected in the research. The results of the extended study may be presented in a more refined way where the gain in the breaths and the number of breaths may be seen as a tool to analyse the different functions in the daily routines of the subjects rather than just going over the testing procedures. The changes in the relevant systems of the human internal system and the physical appearance of the subjects may be closely related to the extended study. The ASR and other type of software may be further elaborated in their use and technical equipment may be used to fine-tune the application of the software rather than implementing the minimum requirements. The microphone may be adjusted further and a variety of microphones may be accounted for in the study. It may further be tested on faster machines which may help in easing out the process of vocal commands (Scott, 1991). Conclusions: The research may be conducted using subjects from age groups that have not been included in the existing study. Furthermore, the subjects may be chosen from many diverse ethnic, social and cultural backgrounds to incorporate factors that are beyond the medical structure into the performance and affects of the vice command systems. Extended research may be done in relation with the frequency of syllables that are present in the breath group along with the types of words used in the research of the subjects. It may however be elaborated by the multi tasking of normal routine where these tests may be run in environments where the subjects are not just using the speech system but also are involved in other tasks such as attending phones and other stress or work increasing activities. On the contrary, environments may also be used where these tests may also be run where there are stress-relieving activities involved. These activities may be defined as the presence of music in the background, scenic views, a pleasant window atmosphere, other people and other forms of comfortable environments. These inputs may also be accumulated in the results and cross examined along with subtle testing environments to make an actual comparison of the stress that is influenced on body functions that may deteriorate health and voice. The outcome of such an extended result may be found to be more relevant to the practical life of many people using voice recognition systems and may as well be tuned into the performance of a voice function as a part of an environment rather than just being a static study into a closed environment. The outcome may be more implied and structured as there will be a diverse nature of subjects and environments and actual results may be more conclusive in relating stress levels that hinder voice health in the long run. Works Cited: Britton, B. K. & Glynn, S. 1989. Computer Writing Environments: Theory, Research, and Design. Erlbaum. Ergonomics Society (ES), 1989. Applied Ergonomics: human factors in technology and society. IPC Science and Technology Press. Huey, B. M. et al. 1993. Workload Transition: Implications for Individual and Team Performance. National Academies Press. Karwowski, W. 2006. International Encyclopedia of Ergonomics and Human Factors. CRC. Panel on Human Factors in the Design of Tactical Display Systems for the Individual Soldier (PHFDTDSIS). 1997. Tactical Display for Soldiers: Human Factors Considerations. National Academies Press. Reid, P. P. et al. (2005). Building a Better Delivery System: A New Engineering/Health Care Partnership. National Academies Press. Robertson, S. 1996. Contemporary Ergonomics 1996. CRC. Scott, D. 1991. Human Computer Interaction. Ellis Horwood. Vincenzi, D. A. et al. 2004. Human Performance, Situation Awareness and Automation: Current Research and Trends. Lawrence Erlbaum. Read More