Investigating an application of speech‐to‐text recognition: a study on visual attention and learning behaviour

Abstract

One major drawback of previous research on speech‐to‐text recognition (STR) is that most findings showing the effectiveness of STR for learning were based upon subjective evidence. Very few studies have used eye‐tracking techniques to investigate visual attention of students on STR‐generated text. Furthermore, not much attention was paid to learning differences, such as learning ability, learning style preferences and gender, to use STR texts. Therefore, this study carried out one experiment. Firstly, participants' visual attention on STR‐generated text was investigated by employing eye‐tracking technique. Secondly, how differently effective STR‐generated texts can be to influence participants' learning achievement was tested. Thirdly, this study compared visual attention and learning behaviour with the different characteristics of participants, such as learning ability, learning style preferences and gender, to use STR‐generated texts. Finally, this study explored students' perceptions regarding usefulness of STR‐generated texts for learning. This paper discusses results, research findings and implications along with conclusions and several suggestions for future development and research.

Innovative Science Educational Neuroscience: Strategies for Engaging Brain Waves in Science Education Research

“…The example of the identification of chemical structural formulas shows that students need to use their spatial and mental rotation abilities to complete the task. Although such abilities could be trained, those talented and gifted students with innate spatial abilities should not be ignored (Liu et al , 2014b;Huang et al 2014;Liu et al 2015). Moreover, how these gifted students use their abilities to identify chemical structural formulas might be important reference material to use in training other students.…”

“…In this chapter, we have mentioned the new trends of combining research on neuroscience and science education, and we also made efforts to do the related researches (Liu et al 2014a;Liu et al 2014b;Huang et al 2014;Liu et al 2015). However, as interdisciplinary researchers, we need to provide information regarding the difficulties faced in the engagement of neuroscience and science education for future researchers.…”

“…The commonly used research approaches used in science education can be divided into quantitative and qualitative approaches (Liu et al , 2014a;Huang et al 2014). In the quantitative research approach, questionnaires and tests are used most, the data from which are analyzed statistically.…”

“…Over the past decade, the interest in brain research has been growing in science education (Huang et al 2014;Liu et al 2015). An increasing amount of evidence from brain research has been applied in the explanation of science educational research, such as the research of Liu and Chang (2014).…”

Research on the Optimal Strategies of Chinese News Program Catalog based on ASR

“…The challenges to extract text information from news programs mainly include mixed voice signal, the length of speech signal (especially for that amplitude is higher than threshold value of continuous time), personalized voice features of newscasts, etc [5,6,7]. For news programs, the paper try to overcome these challenges so as to improve recognition performance by preprocessing, which include speech purified, cutting news programs into sections, reducing of speech speed, voice training, etc.…”