364 C. Zhu and Y. Zhu 5 Experiment and Conclusion Figure 7 shows the waveform of Chinese word “ ” which spell as “ming2yue4” synthesized by PSOLA algorithm only and synthesized by vocal tract cepstrum pa- rameter algorithm overlap on PSOLA algorithm. From the waveform we can see the waveforms between two syllables are different. To test the quality of the word speech sounds produced by the method this paper presents, 30 words were selected. The ex- perimental results showed that the speech synthesized by the new method sounds more natural. Fig. 7. Word’s waveforms synthesized by different methods The quality of the sentence speech sounds produced by the method the paper pre- sents is tested by 30 selected sentences and 8 different people. The experimental results also showed that the speech synthesized by the new method sounds more natural and adapt to the prosody of Chinese poetry. Take the sentence “ / / ” whose spelling is “ju3tou2/wang4/ming2yue4” as example, its waveforms and pitch curves synthesized by two method are shown in figure 8. It is obvious to find that the pitch curve of the speech synthesized by the new method has a gradually decline trend which is more adapted to the Chinese poetry prosody. Fig. 8. Sentence speech synthesized by different methods This paper analyzes the feature of Chinese poetry, and addresses a new method for Chinese speech synthesis which happens to adapted to Chinese poetry’s prosody. The method takes advantages of two basic algorithms, and improves the quality of the synthesized Chinese poetry speech. We can apply the new method of Chinese speech synthesis in our Chinese poetry teaching and learning, because it provides us a more convenient way. Learning Chinese poetry needs reciting, and the reciting needs more A New Chinese Speech Synthesis Method Apply in Chinese Poetry Learning 365 listening, we can reproduction the audio stream of the Chinese poetry through the TTS engine. It only needs to store a few characters which take a small storage spaces on our electronic devices, but provide colorful, varied sounds to us. As the network and wireless technology is developing, it certainly will provide more supports for the Chinese poetry speech learning. References [1] Zhao, Y.: Tone and Intonation in Chinese. Bulletin national Institute of History and Phi- lology Academia Sinica 4(2), 121–134 (1933) [2] Wu, Z., Lin, M.: Experimental phonetics Summary. China Higher Education Press, Beijing (1987) [3] Zhu, Y., Zhang, S.: Initial Consonant and Simple/Compound Vowel Based Chinese-TTS Technology. Modern electronic information technology theory and application, 756–758 (2005) [4] Moulines, E., Charpentier, F.: Pitch-synchronous Waveform Processing Techniques for Text-to-Speech Synthesis Using Diphone Speech Communication, vol. 9, pp. 453–456 (1990) [5] Chen, Y., Zhang, Z.: Application of PSOLA Technique to Chinese Text-to-Speech Con- version System. Computer Engineering 26(1) (2000) [6] Zheng, X., Chai, P.: Analyzing Speech Synthesis Based on FD-PSOLA. Microcomputer Applications 17(7) (2001) [7] Liu, Q., Wang, R.: A new synthesis method based on the LMA vocal tract model. ACTA ACUSTICA 23(3) (1998) F. Li et al. (Eds.): ICWL 2008, LNCS 5145, pp. 366 – 376, 2008. © Springer-Verlag Berlin Heidelberg 2008 Comparing the Impact of Two Different Designs for Online Discussion Yuankun Yao 3132 Lovinger University of Central Missouri Warrensburg, MO 64093 yyao@ucmo.edu Abstract. This study examined the impact of two different designs of online discussion on student participation in the discussions and the type of learning that was generated. The results showed that when students were given an oppor- tunity to respond to each other with the instructor mostly absent in the discus- sion process until the end of the discussions, they would visit the discussion forums more frequently. They were also more likely to pose follow-up ques- tions for their classmates and engage in meaningful inquires. On the other hand, when the instructor posted frequently in the discussion forums, students would respond to him instead of their peers, and the discussions were seldom carried forth in much depth. Keywords: Online discussion; instructional design. 1 Introduction Most instructors of online courses require their students to participate in some sort of online discussions as part of their learning experience. Since online teaching has only been recently introduced in higher education, online instructors have few models to follow in setting up discussions for the students. As a result, instructors may design the discussions in different ways, which could lead to very different learning experi- ences and outcomes for the students. 1.1 Theoretical Rationale for Online Discussion The use of student discussion in an online course is often justified on the learning theory of constructivism, especially social constructivism. According to constructivism when a concept is connected with a learner’s personal experience in an authentic context, learn- ing is more likely to occur [1], [2]. Interaction between different learners is considered important, since it makes it possible for a learner to see how a concept or idea applies to different circumstances [3]. According to Hemetsberger and Reinhardt [4], knowledge- building at the collective level consists in “the advancement and collective elaboration of conceptual artifacts” (p. 190). Dennen and Wieland [5] proposed that when analyzing whether discussion supports learning, one needs to look for learning evidence within a collection of messages instead of examining them in isolation. Comparing the Impact of Two Different Designs for Online Discussion 367 Unlike discussion in a face to face class, discussion in an online environment is of- ten asynchronic [6]. The online instructor and the students do not have to log into the course website at the same time in order to carry on the discussion. As a result, the students have sufficient time to think before responding to a question in a discussion forum [7]. Asynchronic discussion also tends to reduce the participation anxiety of shy students. Possibly for such reasons, online discussions are perceived by instruc- tors as more substantive than discussions in a regular classroom [6]. 1.2 Designs for Online Discussion Although online discussion has the potential to generate collective learning for the students, it does not mean that such learning will easily take place in every online setting [8], [9], [2]. According to Dennen and Wieland [5], students may “nominally fulfill class discussion requirements by posting, but fail to engage more deeply with their classmates by reading and responding to their messages in a thoughtful and re- flective manner” (p. 281). In order for learning to take place, an instructor needs to design the discussions in a way that makes such learning possible. There are different ways in which discussions in an online class may be designed. Some online instructors may let students choose their discussion topics, whereas the majority of instructors like to develop the topics for their students. Even if an instructor is responsible for developing the discussion topic, there are different kinds of topics that may be used. It was found that discussion topics that require students to relate to their experiences are more likely to engage students in the discussions [5], [10]. A related decision in the design of online discussion is the role of an instructor [2], [11]. The instructor has the option to be heavily involved in the discussions, by re- sponding to every student’s comments, or remain as an observant during the process, providing little input until the end of the discussion. It was generally believed that students can have successful online discussions without much intervention from the instructor [5], [10]. According to Dennen [9], when there was considerable presence of the instructor, students would choose to respond to the instructor instead of their peers. In a study by [11], it was reported that the frequency of postings by online instructors had significantly negative correlations with the frequency and length of student postings. The study also found that students tended to consider instructors who were active in posting as more enthusiastic and more knowledgeable, especially if instructors provided postings towards or at the end of the discussions. According to the study, the timing of instructor postings did not have a significant impact on stu- dent participations in the online discussions. 1.3 Purpose of Study This study compared two ways of designing the discussion forums in an online class for teacher candidates. The purpose was to find out if requiring students to respond to one another with little intervention from the instructor would lead to increased student participation and more significant student learning, as compared to a design that re- quired students to directly respond to the original discussion topic followed by an instructor response. 368 Y. Yao 2 Method This study used a mixed method of quantitative and qualitative inquires. The quantita- tive approach was used to collect data regarding the extent of instructor and student participation in online discussions. The participation data include: 1) the number of times the students and the instructor visited each discussion forum; 2) the number of times posted a message in a discussion forum; and 3) the number of words for each posted message. The qualitative approach was utilized to examine both the process and product of the discussions associated with the two designs. To obtain the qualitative data, the researcher chose two discussion forums that used the same topic and similar instruc- tions and compared the discussions in these two forums in terms of: 1) the degree of interactivity between the students; 2) the opportunity for students to raise thought- provoking questions; and 3) the evidence of collective learning for students. 2.1 Context of Study The data for the study came from two online course sections taught by the researcher in Fall 2007 and Spring 2008 in a Midwestern state university in the U.S. The course was “Educational Measurement and Evaluation,” a two credit hour class designed for both undergraduate education majors and graduate students seeking their initial teacher certification through an alternative route. The course was offered completely online using the Blackboard course management system. In both sections of the course, there were 24 active students. There were altogether 14 lessons in the course, with each lesson covered in a week. The data of this study were based on the discus- sions for the first six lessons in both semesters. There were a total of 12 discussion topics from each semester that were examined, with two topics for each lesson. The topics and instructions for discussions are provided in the Appendix. As an instructor of the course, the researcher had access to the course statistics function of the Blackboard to obtain the number of times students visited each discus- sion forum and the number of times they responded in each forum. The researcher also had access to a collection function in each discussion forum that simplified the process of calculating the length and viewing the content of each posting. 2.2 Subjects of Study In the Fall 2007 section, the students were instructed to either directly respond to the instructor’s question, or respond to each other. The instructor responded to most stu- dents’ messages if they were posted by the Thursday deadline. In the Spring 2008 section, the students were asked to respond to their previous respondent only. The instructor waited until the end of the discussion before providing a summary that included comments on some of the ideas students shared and answers to any unan- swered questions. The discussion topics from the two sections were worded in a simi- lar way (Appendix). . as part of their learning experience. Since online teaching has only been recently introduced in higher education, online instructors have few models to follow in setting up discussions for. technology is developing, it certainly will provide more supports for the Chinese poetry speech learning. References [1] Zhao, Y.: Tone and Intonation in Chinese. Bulletin national Institute of History. that when analyzing whether discussion supports learning, one needs to look for learning evidence within a collection of messages instead of examining them in isolation. Comparing the Impact