Developing Visual Novel Game with Speech-Recognition Interactivity to Enhance Students ’ Mastery on English Expressions

The teaching of English-expressions has always been done through conversation samples in form of written texts, audio recordings, and videos. In the meantime, the development of computer-aided learning technology has made autonomous language learning possible. Game, as one of computer-aided learning technology products, can serve as a medium to provide educational contents like that of language teaching and learning. Visual Novel is considered as a conversational game that is suitable to be combined with English-expressions material. Unlike the other click-based interaction Visual Novel Games, the visual novel game in this research implements speech recognition as the interaction trigger. Hence, this paper aims at elaborating how visual novel games are utilized to deliver Englishexpressions with speech recognition command for the interaction. This research used Research and Development (R&D) method with Experimental design through control and experimental groups to measure its effectiveness in enhancing students’ English-expressions mastery. ANOVA was utilized to prove the significant differences between the control and experimental groups. It is expected that the result of this development and experiment can devote benefits to the English teaching and learning, especially on English-expressions.


Introduction
Communication is what language like English is for (Waterson, 2008).For those who believe in behavioral approach, language practices in form of drilling become much of the solution (Richards, 2005).As a result, English expressions like greetings, apologizing, congratulating, and the like are taught for conversation practice, and are considered as real communications (Davies & Pearse, 2000).Thus, Language expressions such those become one part of teaching materials in most academic levels of English curriculum in countries like Indonesia where English serves second language or foreign language.
Based on the observation conducted in some senior high schools, English expressions like greetings, apologizing, and the like have been presented through providing conversation samples in form of printed texts, audio recordings, and videos.
The teachers usually implement conversation practice in which a student is partnering with a peer, demonstrating the sample conversation already served by the English textbooks.In addition, the Computer game, as one form of computer aided learning advancements, is able to motivate students to be involved in learning processes (Liu & Chu, 2010).
The visual novel game in the study of Muhammad (2016)  The interaction used in Muhammad (2016) was click button.As Muhammad's suggestion, future researcher was encouraged to utilize speech recognition as interaction so that the conversation feels like real.Speech recognition has been used for many purposes of interaction.One of simple implementations is that used by Wardhany, et. al. (2014).
Thus, this research aimed at answering the following research question: Do students taught using the Visual Novel Game with speech recognition interactivity have higher achievement on English expression than those taught using conventional method?
In response to the research question, the following hypothesis was proposed: There is significant difference on the students' English expressions mastery between experimental and control group.

Related Works
This research and development is related to some previous studies concerning with computeraided learning in form of games like visual novel for educational purposes in general, and teaching and learning English in particular.Emi & Okuda (2016)

Originality
The idea of utilizing speech recognition as a medium for interactivity in visual novel game has never been implemented before, knowing that the remaining visual novel games are click-based interaction.Thus, the development of the visual novel game in this study stands on its novelty.
Unlike other existing visual novels, the visual novel developed in this study shows the following differences (Table 1).
Speech recognition is utilized as interaction medium, replacing button click as used in the remaining visual novel games.The consideration in using this type of interaction deals with facilitating the role of language as spoken medium of communication.
In real life usage, Englishexpressions are used for communication.Thus, speech recognition as the main interaction will bring the sense of real communication.

System Design
Visual novel game in this study was composed of elements namely (1) visual assets including background and character graphics, (2) audio assets including characters' voice over, background music, and audio effects, (3) branching dialogue, (4) grading system, and (5) interactivity.
The visual assets were composed with manga style, common graphical style for most visual novels.
The voice over audio, recorded from the real voices of native English talents, represents the characters' responses to the player.
The dialog branching was developed within each task of each mission that connects the whole story of the game.The dialog flow was similar to adventure game like choose your own story games in which each response represents different consequence for either successful or unsuccessful.
Each success or fail is recorded as point for grading.
The main story map is shown in Figure 2.
In the game, the player is given response options.The difference between common visual novel game and this visual novel game is that this game uses speechrecognition to trigger the button, while other visual novel games use mouse-click to trigger the button.
The conversation system in which speech-recognition is processed as the interaction is shown in Figure 1.A user gives voice input that is converted as text and is assigned as keyword.If the words match with the pass keywords based on the question by the character, then the button will be activated and answered by the computer through the character.

Results
The effectiveness of visual novel game developed in this research in enhancing students' mastery of English expressions was examined through experimental design.Control and experimental groups were assigned with different treatments.The control group was given the existing learning source, while the experimental group was given the visual novel game as the learning source.
The posttest was conducted to see the increase after each treatment.
The achievement increase of each group was then compared to reveal the difference.The raw test result of both groups is shown in Table 1.The flow of this experiment is elaborated in Table 2.

Research Subjects
The research subjects of this study were two classes in Politeknik Elektronika Negeri Surabaya (PENS), which were assigned as control group with 27 students, and experimental group with 28 students of Multimedia Broadcasting, one of majors in PENS.
The two classes were considered as equal because  3.

Instruments
As this research aimed at examining the effects of the visual novel game towards students' achievement.Achievement test was used to see the students' mastery on English expressions.
The achievement test was the test developed by the Team of English lecturers in PENS.The test consisted of English expression mastery in form of multiple-answer questions.The test was tested for its reliability by the team.The reliability was measured using Anates v.4 for windows through split-half method with Pearson's Product Moment and the reliability score shows 0.79, which is considerably high (Arikunto, 2009).

Achievement Results
To find out whether the experimental group significantly achieved higher score than the control group, the posttest results of both groups were compared.As the number of students in each group was less than 30, ANOVA was implemented to reveal the significant increase.There is a chance for games to be applied as

Suggestion
As a suggestion for further research, it is encouraged that learning material be integrated with games.Fasli & Michalakopoulos (2005) and Gros (2007) claimed that in order to support active learning, games may be utilized as medium to deliver school contents, regardless of their possible obstructions.
In the future, data of students' participation in using this visual novel game can be recorded to measure students' learning activity Rachmawan et.al. (2016).
teachers sometimes encourage the students to develop their own conversation script containing English expressions and then to demonstrate it.The assessment the teachers frequently implemented to evaluate the students' achievement in English expressions was through written test.These practices, in which English expressions are given through written conversation samples, audio recordings, videos; and are assessed through written test, or sometimes through oral test, have been experienced by students across their academic levels from elementary, junior high school, and senior high school.As a result, as what most teachers suspect, students tend to start experiencing boredom of having been through similar practices.Regarding with this problem, the teachers demand another form of teaching medium to deliver English expressions materials to the students.Responding to this expectation, Muhammad (2016) combined computer aided learning technology with educational contents by developing and experimenting a visual novel game associated with conversation gambits containing English expressions.
presents in a form of dialog branches utilizing English expressions and conversation gambits as the options of responses by the player.The game play involves the steps including player selects a mission to accomplish, and follows the dialog branching by choosing offered responses.Each response leads to different results, either successful or failed .The effectiveness of the visual novel game was experimentally assessed through control and experimental groups.It resulted that it brought significant improvement on the achievement but less significant increase on the motivation.
they were distributed based on the results of admission test conducted by PENS.The admission test resulted 55 students, comprising 15 students above average score, 28 students in average score, and 12 students of below average score.They were then assigned to class A and B. Class A consists of 8 students above average score, 14 students in average score, and 6 students of below average score.The remaining numbers were assigned into class B. Treatments Control and experimental group received different treatments.The control group was given lesson on English expressions through conventional method in which the lecturer used printed materials for students to learn English expressions.The experimental group used visual novel with speechrecognition based interaction.The detail of each treatment is shown in Table

P
value and statistical significance: The twotailed P value is less than 0.0001.By conventional criteria, this difference is considered to be statistically significant.Confidence interval: The mean of Experiment minus Control equals 10.85.95% additional medium for students to acquire learning materials other than in classrooms.Hung, Yong & Lin (2015) confirmed that games effectively reduce gaps in terms of students' achievement as outside classroom, students can outperform autonomously.

Table 4 .
Unpaired T-test result

Table 3 .
Experimental Research Design