9 pages due Wednesday 4pm

Purpose

Speech recognition software has existed for decades; different modifications have been made to the application to enable custom uses. The goal of this research is to create a program that will use speech recognition features to help speakers who have trouble with English word’s pronunciation to improve their speech pattern. The program will also help them to track their speech improvement progress and receive a feedback.

Project objectives

To help people with difficulties with English words improve their speech patterns

To help English language learners track their progress in speech improvement

Related Work

1. Speech recognition in assistive technology

Speech recognition is defined as the process through which human words are converted into a format that can be readable in a machine, for example, a computer or even a phone. Speech recognition technology which is also often referred to as computer speech recognition or the automatic speech recognition are software and devices which are used to convert words that are spoken into text(Ifukube, 2017). It works by an individual speaking into the microphone of the device which is connected to a machine or a sound detecting software, and then the sound is converted into numbers and to a readable text through algorithms. This assistive technology plays a very major role in improving speech patterns and more so in education as will be discussed below.

The first way through which speech recognition technologies can be used in assisting to improve the speech patterns is by allowing a greater level of interaction with disabled individuals. Physical disability can be a big barrier to communication between individuals; for example, it is impossible to have a conversation with an individual who does not have the hearing ability. Speech recognition helps to break such barrier and enhance speech and communication by for example by converting the spoken words into a readable text which the other individual can read and also respond. This technology can also help students who do not have motor skills such as the use of their hands or legs to still do the same tasks as the other. They can use the technology to make their reports and research and write just like any other student.

Second, the speech recognition technology has been very instrumental in improving speech patterns by improving the level of fluency of individuals. This is because the speech recognition software has evolved to a level where it has the ability to gauge the reading ability of individuals. Individuals can simply read into the machine, and it can be able to gauge their ability and tell what they are reading right and what they are reading wrong and gauge their fluency levels. An individual can very easily make use of this technology to enhance their reading abilities, their pronunciation and also to practice fluency.

Third, the speech recognition technology is very useful in improving speech patterns by enhancing language development. In this case, language development does not only apply to the children but also for adults. There are very many languages out there that we can learn and the speech recognition software is very instrumental in the learning of new languages. The technology has language software whose function is to check how well an individual can speak a specific language.

For example, if an individual is learning Spanish, they can speak some words to the machine, and the machine can tell how well they are reading these words. Speech recognition also enhances the ability of an individual learning a new language by translating one language to another The software has the ability to convert one language into another, for example, English to Spanish, and this serves as a very effective way of learning another language. Finally, this software enhances language learning through its ability to speak to an individual in a given language, and when the individual responds, the machine evaluates the response to tell whether or not it is correct (Craddock, 2003).

In general, the speech recognition technology helps improve speech patterns for disabled individuals, enhancing language fluency for individuals and also helps in learning new languages. It serves as an assistive technology which can make an individual’s speech abilities better through all these ways.

2. Other related works

The speech recognition software allows for people to talk and words appear. All that one need says a command and the computer obeys. These programs provide speed through documentation. They have a fast dictation and a high recognition accuracy that continually improves as they adapt to one’s voice. They define short-cut repetitive tasks by speeding up documentation creation and boosting productivity. Dragon natural has long ruled the speech recognition landscape. The more choices given to using voice commands the better. The spoken words better suit specific projects than typing. Windows speech recognition by Microsoft has newer versions of speech recognition. It is said to be the most available mainly because it’s being offered for free on Windows 7 or Windows 8 PC. This allows people to open tabs and browse by voice commands. However, the program’s performance can stand improvements especially when it comes accuracy levels.

There are others speech recognition programs such as Google voice and Tazti. However, Dragon Natural speaking far beats the rest. The windows one provides a formidable rival but lacks accuracy. Tazti is more geared towards by gamers who want to use their voice and send characteristics in battles. Users can launch programs, start a media player without the use of a keyboard. However, the program doesn’t use dictation. Others such as Sonix are used to transcribing and editing videos and audios. Overall, dictation programs have come on massive leaps and bounds from not only dictating emails and documents but controlling the PC voice commands.

Technology used in speech therapy (developmental)

Technology is used more in speech-language pathology. Speech-language uses tablets, computers, and smartphones for various purposes. Parents download the apps and programs to help their children practice their speech and language skills at home. Speech-language therapy services have also found its way into the internet through various channels such as Skype. Therapy tools are meant to motivate and open up opportunities for speech and language developments in children. The technology has allowed for articulation focused on words and sentences and story building. Parents may use such technology as part of speech program. This has helped children build up on their language skills.

Students in the department of communicative disorders are pioneering on ways to support children in severe communication needs. Speech developmental needs are mostly used for conditions such as autism, cerebral palsy and pervasive developmental delays. Communication methods are introduced to capitalize on their strengths and attempt to shape new knowledge through a guided learning experience that facilitates participation and learning. Children are allowed to use visualized language in real time through the graphical user interface. Language proficiency is conducted by use of formal and informal assessments undertaken by bilingual educators or speech psychologist. The method involves translating written symbols into speech and their interpretations (Burns et al., 1998). Technology use in speech therapy: rehab from physical injury or stroke (brain injury)

Communication technologies are used in rehabilitation of traumatic brain injury. The major types of technology include assistive technology, augmentative and information communication technology.  The interventions assist the patients in enhancing participation through modified contextual factors. The instructional methods use declarative memory such as imagery and verbal elaboration. Computer-assisted treatment makes use of programs through repetitive trials. The speed and complexity of the exercise increases as the user's performance improves to maintain a high portion of favourable traits. The programs are used in conjunct to clinical-moderated treatment.

Patient-centred design of cognitive assisted technology assists in a range of computer aided gait. High skilled therapists are adapted to the use of technology for optimum recovery. The sophisticated machines blend in with the experts knows how. A process designed for maximum gains. Cognitive rehabilitation uses the multi-disciplinary approach with neuropsychiatric evaluations. Significant improvements are attained by undergoing direct attention in speech training. Brain-injured persons may suffer from ideomotor, ideational and constructional apraxia. The cognitive communication disorder impairs social interaction. The principle use of the technology-aided guide is to mass practice language tasks of increased difficulty and using constraints of non-verbal communication.

Proposed Approach

The nature of problem calls for a Quantitative experimental approach in realizing the goals of this research’s. Having identified the problem, scientific hypothesis need to be developed and effective experimentations done. In addition, the work will entail a lot of program development and testing in a bid to realize the scope of the research. After developing or identifying the correct programs, a population sample will be selected for purpose of study for a given duration. Feedback will be collected and analysis, recommendations and conclusions made.

Required resources

The nature of research calls for the following:

Related researches for reference

Program developer

Population sample containing English language learners with access to computers and developed software.

References

Burns, R. B., Crislip, D., Daviou, P., Temkin, A., Vesmarovich, S., Anshutz, J., ... & Jones, M. L. (1998). Using telerehabilitation to support assistive technology. Assistive Technology, 10(2), 126-133.

Carey B. (2016). Smartphone speech recognition can write text messages three times faster than human typing. Stanford News. Accessed from: http://news.stanford.edu/2016/08/24/stanford-study-speech-recognition-faster-texting/

Julie P & Zeanah E. (2005). An evaluation of voice recognition software for use in interview-based research: a research note. Sage Journals. http://journals.sagepub.com/doi/abs/10.1177/1468794105050837

Craddock, G. (2003). Assistive technology: shaping the future; AAATE'03. Amsterdam Tokyo: IOS Press Ohmsha.

Ifukube, T. (2017). Sound-based assistive technology: support to hearing, speaking and seeing. Cham, Switzerland: Springer.