2017 IEEE 17th International Conference on Advanced Learning Technologies
Applying speech-to-text recognition and computer-aided translation for supporting multi-lingual communications in cross-cultural learning project Rustam Shadiev1, Barry Lee Reynolds2, Yueh-Min Huang3, Narzikul Shadiev4, Wei Wang1, Rai Laxmisha5, and Wanwisa Wannapipat6 1
Nanjing Normal University, Nanjing, China 2 University of Macau, Macau, China 3 National Cheng Kung University, Tainan, Taiwan 4 Samarkand State University, Samarkand, Uzbekistan 5 Shandong University of Science and Technology, Qingdao, China 6 Khon Kaen University, Khon Kaen, Thailand Email:
[email protected],
[email protected],
[email protected],
[email protected],
[email protected],
[email protected] programs (Ertmer et al., 2011). Nevertheless, some issues still exist, which hamper intercultural interactions and exchanges among people. According to Osman and Herring (2007), Reynolds (2013, 2015), and Shadiev and Huang (2016), the language barrier is the most critical factor in cross-cultural learning as it makes difficult to communicate among people without language in common. To address this issue, we applied speech-to-text recognition or STR (Hwang, Shadiev, Kuo, & Chen, 2012; Kuo, Shadiev, Hwang, & Chen, 2012; Shadiev, Huang, & Hwang, 2017) and computer-aided translation or CAT (Shadiev & Huang, 2016) systems in this present study. We aimed (1) to measure accuracy rate of STR and CAT technologies and (2) to investigate whether our approach is feasible to facilitate cross-cultural learning.
Abstract— We applied a speech-to-text recognition (STR) and computer-aided translation (CAT) systems to support multilingual communications students participating in crosscultural learning project. The participants were engaged in interactions and information exchanges in order to learn and understand cultures and traditions of their peers. Their communications were carried out in their native languages on social communication platforms. The participants spoke and STR system generated texts from their voice inputs. CAT system then simultaneously translated STR-texts into English. Finally, translated texts were posted on social communication platforms along with spoken content in the participants’ native languages. We aimed to examine accuracy rates of processes associated with STR and CAT for different languages during multi-lingual communications in our cross-cultural learning project. In addition, the feasibility of our approach to support multi-lingual communications in cross-cultural learning project was investigated. Our results showed that the lowest accuracy rate was for Mongolian and Filipino and the highest was for Spanish, Russian, and French. Our results also demonstrated that cross-cultural learning took place; the participants understood and were able to explain foreign traditions to others as well as to compare foreign traditions with their own local. Based on our results, we made several suggestions and implications for the teaching and research community.
II.
Twenty one university students participated in this study. There were representatives of thirteen nationalities. Cross-cultural learning project included four one-week steps: (1) the participants introduced themselves, their hobbies and interests; (2) the participants introduced their local traditions and related culture; (3) each participant selected one tradition, experienced it and shared his/her personal experience of selected tradition with other participants; (4) all participants met online face to face to communicate synchronously with each other about themselves, their traditions, and experiences of foreign cultures. The participants communicated via Facebook during the first three steps and via Skype in the fourth step. The participants used STR technology to generate texts from their voice input in their native languages and then they used CAT technology to simultaneously translate STR-texts into English. We employed Google Translate system for STR and CAT processes. The data for analysis was collected from two main sources: online communication among the participants and one-on-one semi-structured interviews following general
Keywords- Speech-to-text recognition, Computer-aided translation, Cross-cultural learning, Accuracy rate.
I.
INTRODUCTION
Because cultural differences exist in our globalized society, cross-cultural understanding became an important issue nowadays (Bentley, Tinney, & Chia, 2005). One reason is that people need to understand cultural differences in order to avoid any problems during intercultural communication and co-exist with others peacefully in this interconnected world (Rogers, Graham, & Mayes, 2007). Modern information and communication technologies play important role in aiding cross-cultural learning 2161-377X/17 $31.00 © 2017 IEEE DOI 10.1109/ICALT.2017.20
METHOD
182
languages. This approach can be used for educational programs related to cross-cultural learning.
recommendations from earlier studies (Shadiev, Hwang, & Huang, 2017; Shadiev, Wu, & Huang, 2017). III.
ACKNOWLEDGEMENT
RESULTS AND DISCUSSION
This research was partially supported by the project of National Education Science Foundation of China (BCA150054).
A. Accuracy rate According to our results, the lowest accuracy rate was for Mongolian (94.37%) and Filipino (94.60%) and the highest rate was for Spanish (98.15%), Russian (98.02%), and French (97.95%). Perhaps, this is due to wide use of Spanish, Russian, and French in contrast to Mongolian and Filipino. CAT database is bigger for Spanish, Russian, and French and smaller for Mongolian and Filipino. With small language database CAT translates texts with lower accuracy rate and with large database CAT translates texts with higher accuracy rate. Another reason for different accuracy rate is due to the similarities or differences between the English and these languages; the larger difference, the lower accuracy is (Tobin, 2015). Mongolian and Filipino are Asian languages and has larger differences with the English compared to the European languages, such as Spanish, Russian, and French.
REFERENCES [1]
[2]
[3]
[4]
B. Facilitating cross-cultural understanding We asked the participants in the beginning of this study to select and experience the culture they are not familiar with. So the participants admitted that before our crosscultural learning activity they had no prior knowledge regarding the culture and traditions they experienced. After the cross-cultural learning activity we assessed cognitive level of the participants and results showed that all participants reached “Understand” levels. That is, the participants understood foreign traditions they experienced in our learning activity. According to our results, the participants could recall, interpret, summarize, compare and explain traditions they experienced (Anderson & Krathwohl, 2001). We interviewed the participants to explore their perceptions regarding cross-cultural learning activity supported by our technological approach. In the interviews, the participants mentioned that this approach was interesting and useful for their cross-cultural learning. IV.
[5]
[6]
[7]
[8]
[9]
[10]
CONCLUSIONS
Our results suggest that using our approach is useful for daily life simple communication in most languages. However, when considering communication on complex and advanced topics, our approach for widespread languages and languages similar to the English (e.g. Russian, French, and Spanish) produce more accurate content. At this moment, STR and CAT should not be considered as a well-rounded professional translation mechanism from voice input as it has its own limitations that need to be considered. Our results can be useful for educators and researcher who are going to apply STR and CAT for various educational and research projects in the near future. Based on our results, we suggest to apply similar approach for supporting cross-cultural communication among students speaking in different foreign
[11]
[12]
[13]
[14]
183
Anderson, L. W., & Krathwohl, D. R. (Eds.). (2001). A taxonomy for learning, teaching and assessing: A revision of Bloom’s taxonomy of educational objectives. New York: Addison, Wesley, Longman. Bentley, J.P.H., Tinney, M.V., & Chia, B.H. (2005). Intercultural Internet-based learning: Know your audience and what it values. Educational Technology Research and Development, 53, 117-127. doi:10.1007/BF02504870 Ertmer, P.A., Newby, T.J., Liu, W. et al. (2011). Students’ confidence and perceived value for participating in cross-cultural wiki-based collaborations. Educational Technology Research and Development, 59, 213-228. doi:10.1007/s11423-011-9187-4 Hwang, W. Y., Shadiev, R., Kuo, T. C. T., & Chen, N. S. (2012). A study of speech to text recognition and its effect to synchronous learning. Journal of Educational Technology & Society, 15(1), 367– 380. Kuo, T. C. T., Shadiev, R., Hwang, W. Y., & Chen, N. S. (2012). Effects of applying STR for group learning activities on learning performance in a synchronous cyber classroom. Computers & Education, 58(1), 600–608. Osman, G. & Herring, S.C. (2007). Interaction, facilitation, and deep learning in cross-cultural chat: A case study. The Internet and Higher Education, 10(2), 125-141. Reynolds, B.L. (2015). Helping Taiwanese graduate students help themselves: Applying corpora to industrial management English as a foreign language academic reading and writing. Computers in the Schools, 32(3-4), 300-317. Reynolds, B.L. (2013). A web-based ELF writing environment as a bridge between academic advisers and junior researchers: A pilot study. British Journal of Educational Technology, 44(3), E77-E80. Rogers, P.C., Graham, C.R. & Mayes, C.T. (2007). Cultural competence and instructional design: Exploration research into the delivery of online instruction cross-culturally. Educational Technology Research and Development, 55, 197-217. doi:10.1007/s11423-007-9033-x Shadiev, R. & Huang, Y.M. (2016). Facilitating cross-cultural understanding with learning activities supported by speech-to-text recognition and computer-aided translation. Computers & Education, 98, 130-141. Shadiev, R., Huang, Y.M., & Hwang, J.P. (2017). Investigating the effectiveness of speech-to-text recognition application on learning performance, attention, and meditation. Educational Technology Research and Development, DOI:10.1007/s11423-017-9516-3. Shadiev, R., Hwang, W.Y., & Huang, Y.M (2017). Review of research on mobile language learning in authentic environments. Computer Assisted Language Learning. http://dx.doi.org/10.1080/09588221.2017.1308383 Shadiev, R., Wu, T.T., & Huang, Y.M. (2017). Enhancing learning performance, attention, and meditation using a speech-to-text recognition application: evidence from multiple data sources. Interactive Learning Environments, 25(2), 249-261. Tobin, A. (2015). Is Google Translate Good Enough for Commercial Websites?: A Machine Translation evaluation of text from English websites into four different languages. Reitaku Review, 21, 94-116.
