An Experimental Study of Chinese Disyllabic Tone Errors in Native French Speakers and an Investigation of Online Teaching Strategies

: In the learning of Chinese as a second language, phonetic learning serves as the basis of the whole learning process, among which the learning of Chinese tones is of paramount importance. In this regard, this paper aims to explore the pronunciation of Chinese tones by native French speakers in the target language environment. First, through the experimental analysis as well as the recording of the tones of native French speakers, this paper selects a total of 237 disyllabic words, with a total of four combinations determined except for the combination of the same disyllabic neutral tone in the target language environment. Meanwhile, this paper utilizes PRAAT software to acquire the average frequency of tone sampling points, whereby the tone-pattern error of each tone combination can be clarified. Secondly, by comparing and analyzing the foregoing results with the tone patterns of Chinese students, this paper further finds out the tone-pattern errors of the same tone combination in disyllabic words, thus summarizing the corresponding error rules. Through the above experimental analysis, this paper concludes that the combination of Tone 3 and Tone 3 presents the highest pronunciation error rate, followed by the combination of Tone 2 and Tone 2. In addition, this paper reveals that the tone error is the main tone error and that there are also duration errors. Lastly, combined with the specific reasons for the errors, this paper proposes targeted online teaching strategies suitable for international Chinese education.


Introduction
Along with the continuous strengthening of China's economic strength, an increasing number of foreigners begin to learn Chinese, using it as a second language after their mother tongue. Yet, the problems related to speaking Chinese with a foreign accent are still inevitably exposed in Chinese learning. Against this background, how to better solve the problems related to learners' pronunciation and intonation has become a top priority. To this end, this paper aims to identify relevant tones that are most likely to lead to errors by Chinese learners through experimental research, thereby proposing targeted online teaching strategies suitable for international Chinese education.

Domestic Research on Chinese Disyllabic Tone Errors
Through the tone test of disyllabic words for six Japanese students who have studied Chinese for one and a half years, Liu (2008) found the tone combination of Tone 2+Tone 2 with the highest error rate, followed by Tone 3+Tone 2. a clnixster@126.com, *cicsbl@126.com Among them, the error of Tone 3+Tone 2 is expressed as Tone 2+Tone 3 [2] . It can be seen that Japanese students' tone errors are mainly concentrated in Tone 2 and Tone 3. Moreover, based on the perception experiment concerning the tones of disyllabic words conducted among 15 freshmen in the Chinese Department of Cairo University, Ji (2010) concluded that the tone combination of Tone 3+ neutral tone presented the highest error rate, followed by that of Tone 3+Tone 2 [1] . Xin (2011) selected 10 Italians, encompassing 5 males and 5 females, as experimental subjects, indicating that the tone combination of Tone 3+Tone 2 presented the highest error rate, followed by Tone 2+Tone 1 [4] . Through the tone sampling experiment of 30 Central Asian students, Zheng (2020) compared and analyzed the experimental results with PRAAT software, demonstrating that Tone 2 and Tone 3 had the most error rates. As outlined above, by virtue of these sets of data, this research found that Tone 2 and Tone 3 are the vocal difficulties of countries without tone language [7] .
On the other hand, the reason why this research chose to explore the tone errors of native French speakers lies in the profound friendship and historical precipitation between China and France. Meanwhile, on the occasion of the 60th anniversary of the establishment of diplomatic relations between China and France, the proportion of native French speakers learning Chinese is on the rise. This paper, therefore, intends to take this as the research objective to launch a systematic exploration.

Experimental Subjects
The author conducts experimental research on the tones of 7 subjects whose mother tongue is French in the target language environment. On the whole, the experimental subjects are 7 foreign students whose mother tongue is French (including 4 French, 2 Quebec Canadians, and 1 Malagasy), aged between 20 and 35. Of these, six foreign students are studying Chinese in Classes A, B, and C of the language class of South China Normal University, while one foreign student is studying Chinese in Class A of the language class of the College of Chinese Language and Culture, Jinan University. All of the aforesaid subjects have more than half a year's Chinese learning experience. As shown in the following table, the subjects consist of five females and two males in total, with F representing the female and M representing the male. (Table 1)

Experimental Materials
The HUAYU YUYIN edited by Sun Yuqing is selected as the experimental material, with 237 vocabularies involving 474 Chinese characters being tested [3] .

Experimental Apparatus
In terms of the experimental apparatus, PRAAT speech analysis software is used for monophonic recording with a sampling frequency of 22050 HZ. The recording is saved in a wav-based format. Likewise, PRAAT software is used to perform corresponding speech analysis on the recorded data.

Experimental Procedure
Step 1: After finding the experimental subjects and arranging them in a quiet classroom, the author will distribute relevant experimental materials to these foreign students so that they can get familiar with them in advance. For foreign students in Class A, the preparation time will be extended. Regarding difficult pronunciations, the author will provide subjects with exemplary pronunciations in advance. Each student has 10 minutes to prepare. During the test, tone errors, rather than initial-consonant and compound-vowel errors, are solely taken into account; Step 2: Experimental data will be collected; Step 3: The data will be imported into PRAAT software for acoustic analysis; Step 4: The sonogram of disyllabic words will be obtained for comparative analysis with pronunciation and standard sonograms of Chinese students; Step 5: Conclusions will be drawn.

Experimental Analysis of the Errors in Tone
Combination of Tone 1+Tone 1 As can be seen from the data, the starting frequency of female pronunciation is at least above 220 HZ, whereas that of male pronunciation is above 140 HZ (Table 2), and their pronunciation of first-syllable Tone 1 is obviously not as stable as that of second-syllable Tone 1. From the fundamental frequency, it can be seen that the secondsyllable pronunciation showcases a small range of change, with each subject remaining stable at a numerical value. The reasons for the unsteadiness of the first-syllable pronunciation can be summarized into two points, one of which is the influence of breath, while the other is the failure of the subjects to fully grasp the tone. Additionally, from the perspective of the sonogram, based on the comparison of the pronunciation of Tone 1+Tone 1 between foreign students and China students(Figure1, Figure2, Figure3), it is indicated that the pronunciation frequency of foreign students is fluctuating, which is not like the level tone of China students. Affected by the speech flow, the first-syllable ending tone of foreign students decreased, while their second-syllable beginning tone of them increased. Though males' tone amplitude was lower than that of females, male foreign students generally showed a low tone in terms of Tone 1. Secondly, regarding duration, some foreign students' first-syllable pronunciation was kept at 190 MS, while their secondsyllable pronunciation was kept at 490 MS. The excessively significant gap reflected by the above phenomenon makes their pronunciation sound obviously unnatural. In contrast, some female foreign students had an obvious rising tone in the ending tone regarding the pronunciation of the second-syllable Tone 1. Obviously, this is beyond the pronunciation range of level tone.  From the above fundamental frequency data, it can be seen that among the 11 sampling points related to the second-syllable Tone 2, half of the sampling points are substantially the same in frequency value, with the amplitude varying in a fairly small range of 1-3 HZ. Compared with Tone 3 and Tone 2 35, the subjects made a tone of 33 in the rising tone of 3-5 tone pitch. The pronunciations of subjects F1, F2, and F4 were relatively standard. Subject F3 pronounced part of the first-syllable Tone 2 as Tone 1, and the tone of her second-syllable Tone 2 did not change significantly, showing the amplitude of decreasing first and then increasing. The second-syllable Tone 2 of subject F5 reflected the characteristics of a concave figure, with her tone pattern close to that of Tone 3. The first-syllable Tone 2 and the second-syllable Tone 2 of subject M1 showed excessively small changes in a rising tone, failing to reach the tone of 35. Subject M2 pronounced first-syllable Tone 2 as Tone 1 and secondsyllable Tone 2 as Tone 4. From the tone-pattern perspective, two subjects had the pronunciation of Tone 1 on first-syllable Tone 2, while only one subject had the pronunciation of second-syllable Tone 2 close to that of Tone 3. (Table 3) On the other hand, most subjects' first-syllable and second-syllable Tone 2 had no significant difference in duration (Figure 4). Only one subject had a shorter duration of first-syllable Tone 2 (i.e., 140 MS) and a longer duration of second-syllable Tone 2 (i.e., 340 MS) in all pronunciations of Tone 2+Tone 2, reflecting a significant difference. We can compared with Chinese student's pronunciation( Figure 5).  Tone sandhi occurs in tone combinations of Tone 3+Tone 3 in pronunciation rules, with the ending pronunciation of Tone 2+Tone 3( Figure 8). As can be seen from the above data, subject F1 pronounced well without significant error. Subjects F2 and F3 pronounced firstsyllable Tone 3 as full Tone 3 with no tone sandhi and pronounced second-syllable Tone 3 correctly. Subject F4 pronounced the first-syllable Tone 3 as full Tone 3 without the tone sandhi of Tone 2, presenting an overly long leveltone time between the falling and rising parts of the second-syllable Tone 3. Subject F5 correctly pronounced first-syllable Tone 3, but pronounced second-syllable Tone 3 as Tone 2. Subject M1 pronounced first-syllable Tone 3 as Tone 1. On the same note, from the above data, the second-syllable Tone 3 did not reflect significant tonepattern ascending and descending changes. Subject M2 pronounced most of the first-syllable Tone 3 as Tone 1, with his second-syllable Tone 3 in the same tone pronunciation as that of M1. Hence, the ascending and descending changes of the tone cannot be distinguished. In addition, from the above data, it can be seen that most of the foreign students tested did not veritably master the rules of first-syllable tone sandhi of Tone 3+Tone 3 (Table  4). Overall, the subjects pronounced most of the first syllable as the full Tone 3, and some of the second-syllable Tone 3 as Tone 2, with some Tone 3 pronunciation being insignificant( Figure 6, Figure 7). On the other hand, regarding the duration, the duration of first-syllable Tone 3 was in the range of 200 MS to 300 MS, whereas that of second-syllable Tone 3 was in the range of 360 MS to 400 MS, with little difference.  Regarding the pronunciation of Tone 4+Tone 4, the recorded foreign students showed the most standard pronunciation in this tone combination (Figure 9,  Figure10). On the whole, the starting frequency of females ranged from 217 HZ and 307 HZ, while that of males was lower, ranging from 150 HZ and 177 HZ( Table 5). The test results of the other six subjects were close to the standard pronunciation, except that subject F5 did not pronounce the first-syllable Tone 4 with a significant decrease in tone. Regarding the decline of the subject's pronunciation, her pronunciation showed obvious errors because she immediately issued a second falling tone before she finished her pronunciation. It is worth acknowledging that this is an individual phenomenon and not universal. In other words, errors may be related to students' phonological knowledge and other factors. As can be seen from the amplitude, the subject's secondsyllable Tone 4 was significantly lower than first-syllable Tone 4 by approximately 50 HZ. According to the standard pronunciation of Tone 4, the starting changes of the two pronunciations should be the same.

Experimental Analysis of the Errors in Tone Combination of Tone 4+Tone 4
Regarding the duration, the pronunciation duration of the subjects was all in the range of 100 MS to 200 MS, which was about the same as the pronunciation duration of Chinese students. Compared with other tone combinations, however, the pronunciation duration of this tone combination is shorter.

Summary of Errors in the Same Tone of Chinese Disyllabic Words
In the case of four groups of two same tone combinations, the tone combination of Tone 4+Tone 4 presents the highest pronunciation accuracy. The tone with the lowest pronunciation accuracy is the first-syllable Tone 3 in the tone combination of Tone 3+Tone 3, with the errors mainly concentrated in the tone sandhi of the first-syllable Tone 3. Consequently, Tone 3 has consistently been a difficult point for foreign students to pronounce, followed by Tone 2.
The pronunciation error of Tone 1 is mainly manifested in the instability of pronunciation airflow. According to the frequency, it can be judged that the pronunciation of Tone 1 is fluctuant, without reaching the pronunciation of tone pitch 55. Additionally, the pronunciation error of Tone 2 is mainly manifested in the following two points. In detail, the first point is that some subjects pronounce Tone 2 as Tone 1, and the rise of the second-syllable Tone 2 is not obvious. The second point can be summarized as the confusion between Tone 2 and Tone 3 in pronunciation. Furthermore, the pronunciation error of Tone 3 is mainly manifested in the absence of tone sandhi of the first-syllable Tone 3. The pronunciation of the subjects was generally Tone 3 with a tone pitch of 214. Meanwhile, there have also been cases in which the second-syllable Tone 3 is pronounced as Tone 2.
For both Tone 3 and Tone 2, subjects had long-lasting level tones in their pronunciations, especially when followed by Tone 2 and Tone 3. The biggest reason for this pronunciation phenomenon is that the subjects are not skilled in these two tones, thus leading them to expect to pronounce these two tones with a buffer period. Most subjects do not have a clear grasp of the pitch changes of 35 and 214 regarding their pronunciation position. This in turn reveals why foreign students are not clear about where the pronunciation key points of Tone 2 and Tone 3 will rise or fall.

Tone Teaching Strategies Based on Online Environment
Wu (2022) mentioned that the worldwide COVID-19 pandemic has brought subversive influence to the traditional teaching mode and concept [5] . Affected by this, a host of online teaching platforms, encompassing Tencent Meeting, Zoom, Google Classroom, DingTalk, Rain Classroom, Superstar Learning, Welink, etc., have evolved into frequently-used teaching tools. Likewise, numerous social platforms, such as WeChat, Tik Tok, Auto Quicker, Twitter, Facebook, and YouTube, have been used to assist Chinese teaching.

Teaching Strategy of Online Flipped Classroom Based on MOOC
Among the various courses of Chinese University MOOC, it is worth mentioning a course called Teaching Related to the Key and Difficult Points of Elementary Chinese Pronunciation. Given that phonetic teaching has constantly been the focus and difficulty in teaching Chinese as a foreign language, this course includes learning concerning simple final, compound final, nasal final, tone, and initial consonant. Furthermore, the flipped classroom, which emphasizes students' dominant position during the learning process, puts forward that the traditional "classroom characterized by teachers' teaching and students' learning" should be transformed into a "teaching mode characterized by students' autonomous learning before class, answering questions, reporting, and discussing learning contents through teacher-student interaction in class".
In essence, the learning of Chinese tones reflects a particularity, that is, the majority of foreigners have not come into contact with tone-based language. Therefore, the learning of tones during the learning of Chinese has evolved into a difficult point and focus. Based on the difficulty of tone learning, it is strongly recommended to design an online flipped classroom based on MOOC course learning. In this way, on the one hand, students can learn the four tones of Mandarin in advance in the MOOC class and record their pronunciations for presentation in the online class. On the other hand, teachers can provide students with targeted exercises and corrections based on their pronunciation until the students make the correct sound, thereby urging them to continue to practice after class.

Teaching Strategies Based on the Use of Record-broadcast Courses
The characteristics of record-broadcast courses are beneficial to repeated viewing, infinite loops, and pausing at any time, without worrying that learners are incapable of keeping up with the learning progress. Concurrently, the production of record-broadcast courses is relatively easy. Moreover, record-broadcast courses are relatively friendly to students who are learning phonetics in the primary stage. Phonetic learning, as an introductory course in Chinese learning, is of paramount importance. In contrast, tone learning is more difficult for most students from non-tonal language countries. Consequently, the application of record-broadcast courses helps students to watch and practice repeatedly, especially for Tone 3 and Tone 2.

Teaching Strategy Based on the Use of Live Class
Wen (2020) mentioned that online live teaching, which is similar to face-to-face teaching in offline classrooms, is characterized by teachers being responsible for online knowledge teaching, while learners are responsible for watching the live broadcast and interacting with teachers at any time by means of audio [6] . Nowadays, this teaching method has become a main form of online teaching in various institutions. In addition to being similar to offline courses, the live class reflects the distinct advantage of using the rich teaching resources of the network to interact with students in real time. In this connection, during tone teaching, especially in the learning of Tone 3 and Tone 2, which are relatively difficult to master, teachers can prepare online solitaire games and crossword puzzles, and show them through animation or idioms. Also, teachers can encourage students to find materials for tone practice, making full use of online resources to arouse students' enthusiasm.
In the teaching of live classes, however, teachers should also be prepared to deal with a series of unexpected situations such as network disconnection and network jam, and make corresponding pre-arranged plans.

Teaching Strategy Based on the Use of "Chinese Plus"
"Chinese Plus" is a non-profit cooperative organization jointly sponsored by more than 20 units represented by Continental Han Feng Network Technology (Beijing) Co., Ltd., Chinese Testing International Co., Ltd., and Alibaba DingTalk to fight the COVID-19 pandemic on March 26, 2020. As a whole, Chinese Plus provides three sections, including the compilation of lecture notes, the compilation of supplementary teaching materials, and the compilation of teaching materials, to show the materials used by teachers in actual teaching. According to the current situation inquired by the author, there are no teachers who have written related lecture notes, supplementary teaching materials, or teaching materials on the Chinese Plus platform. Hence, it is imperative to advocate for teachers to highly value the tone teaching, and proactively engage in the tone-related compilation of lecture notes, supplementary teaching materials, or teaching materials, intending to make full use of the Chinese Plus platform.

Online Teaching Strategy Based on the Use of Online Platforms Such as Tencent Meeting, ZOOM, or WeChat
In an article, Wen (2020) mentioned that the Goethe Institute Peking solicited from the learners, through public recruitment, proposals relating to the use of the ZOOM online meeting space to perform online communication projects. The topics, forms, and invited guests of the activities involved were independently planned and implemented by the project applicants or organizers. Similarly, Chinese learning can also make full use of various ways such as Tencent Meeting, ZOOM, or WeChat groups to establish online learning space. In doing so, the Chinese teacher can also promote the discussion on a topic as a moderator. Taking tone learning as an example, Chinese teachers can use songs, tongue twisters, ancient poems, news, and other means to enable learners to practice tone pronunciation, thereby exercising their oral Chinese expression ability.