In this paper we report on the ASR-based CALL system DISCO: Development and Integration of Speech technology into COurseware for languagelearning. The DISCO system automatically detects pronunciation and grammar erro...
详细信息
ISBN:
(纸本)9781622767595
In this paper we report on the ASR-based CALL system DISCO: Development and Integration of Speech technology into COurseware for languagelearning. The DISCO system automatically detects pronunciation and grammar errors in Dutch L2 speaking and generates appropriate, detailed feedback on the errors detected. We briefly introduce DISCO and present the results of a first evaluation of the complete system.
Second language learners of British English (BE) are typically trained to learn four intonation classes - Glide-up, Glide-down, Dive and Take-off. We predict the intonation class in a learner's utterance by modeli...
详细信息
Second language learners of British English (BE) are typically trained to learn four intonation classes - Glide-up, Glide-down, Dive and Take-off. We predict the intonation class in a learner's utterance by modeling the temporal dependencies in the pitch patterns with gated recurrent unit (GRU) networks. For these, we pre-train the GRU network using a set of synthesized pitch patterns representing each intonation class. For the synthesis, we propose to obtain pitch patterns from the tone sequences representing each intonation class obtained from domain knowledge. Experiments are conducted on speech data collected from experts in a spoken English training material for teaching BE intonation. The absolute improvements in the unweighted average recall (UAR) using the proposed scheme with pre-training are found to be 4.14% and 6.01% respectively over the proposed approach without pre-training and the baseline scheme that uses hidden Markov models (HMMs).
We present a tool for the automatic generation of questions from texts for ESL teaching, using resources and methods from the Natural language Processing (NLP) field. The approach presented in this paper is based on s...
详细信息
ISBN:
(纸本)9781665409568
We present a tool for the automatic generation of questions from texts for ESL teaching, using resources and methods from the Natural language Processing (NLP) field. The approach presented in this paper is based on symbolic methods, using hand-crafted rules. The tool uses linguistic information, such as semantic roles, co-references and named entities, to generate questions and answers from a text selected by a teacher. The generated questions are ranked in order to offer the teacher a varied set to create an exercise. All the elements can be edited in case any error is produced, since NLP tools and methods are not completely accurate. When solving an exercise, the students are given immediate feedback based on the expected answers. A primary evaluation showed promising results, especially for "what" and "who" questions.
The rise of Facebook presents new challenges for matching users with content of their preferences. In this way, the educational aspect of Facebook is accentuated. In order to emphasize the educational usage of Faceboo...
详细信息
ISBN:
(纸本)9783642327896;9783642327902
The rise of Facebook presents new challenges for matching users with content of their preferences. In this way, the educational aspect of Facebook is accentuated. In order to emphasize the educational usage of Facebook, we implemented an educational application, which is addressed to Greek users who want to learn the Conditionals grammatical structure in Filipino and vice versa. Given that educational applications are targeted to a heterogeneous group of people, user adaptation and individualization are promoted. Hence, we incorporated a student modeling component, which retrieves data from the user's Facebook profile and from a preliminary test to create a personalized learning profile. Furthermore, the system provides advice to each user, adapted to his/her knowledge level. To illustrate the modeling component, we presented a prototype Facebook application. Finally, this study indicates that the wider adoption of Facebook as an educational tool can further benefit from the user modeling component.
It is generally acknowledged that collocations in the sense of idiosyncratic word cooccurrences are a challenge in the context of second language learn- ing. Learners often produce "ungrammatical" combinatio...
详细信息
It is generally acknowledged that collocations in the sense of idiosyncratic word cooccurrences are a challenge in the context of second language learn- ing. Learners often produce "ungrammatical" combinations such as *give a suggestion or *make a walk. Advanced computational tools able to aid L2 learners with collocations are thus highly desirable. However, state-of-the- art "collocation checkers" are merely able to detect a possible miscollocation and offer as correction suggestion a list of collocations of the given keyword whose semantics is often ignored. In order to address these shortcomings we propose techniques for collocation retrieval and semantic classification that retrieve, for a given base and the intended meaning, the actual collocate lexeme(s), and techniques for collocation error detection and classification. Given the small size of our learner corpora, we also provide an algorithm for the generation of an artificial collocation error corpus for Spanish.
Processing instruction (PI) is a language teaching technique based upon the model of input processing developed by VanPatten (1993, 1996, 2002, 2004). The present study investigated the effects of PI as well as two ot...
详细信息
Processing instruction (PI) is a language teaching technique based upon the model of input processing developed by VanPatten (1993, 1996, 2002, 2004). The present study investigated the effects of PI as well as two other experimental conditions (traditional instruction and control) on the acquisition of the Spanish copulae ser and estar by 66 Chinese university students enrolled in a blended (partially presential and partially online) fourth-semester language course. The PI treatment condition included non-paradigmatic grammar explanations, processing strategies designed to help learners avoid commiting errors they may be predisposed to make, and structured input activities which eliminate redundant features of language that may make difficult the establishment of form-meaning connections. The traditional instruction (TI) treatment condition included paradigmatic explanations of grammar as well as production-based activities and exercises. Participants assigned to the control condition did not receive any instruction during the course of the study. The results of an experiment comparing the effects of each treatment condition on learners' scores on the Knowledge Test of Spanish Copulae, a measurement instrument designed for this study, found PI to be superior to TI at the immediate posttest level for tasks of interpretation and production. However, mean difference scores for the PI group were not significant when compared to those of the control group. Moreover, the learning gains exhibited by the PI group at immediate posttest were not durative, as they were not significant at the delayed posttest.
In recent years, various innovations have been implemented to assist second or foreign language learners with vocabulary acquisition, including CALL-based tools. As CALL-based tools are widespread, the question arises...
详细信息
In recent years, various innovations have been implemented to assist second or foreign language learners with vocabulary acquisition, including CALL-based tools. As CALL-based tools are widespread, the question arises as to whether they are effective, and if so, whether they are similarly effective for students of different learning styles. The present study has investigated whether Dutch secondary school students with different learning styles (visual versus non-visual learners) benefit from using the popular CALL programme WRTS to learn new English words. Thirty-One Dutch middle school students participated in this study. They were assigned to either the experimental group or the control group. Both groups underwent pre-testing, training and post-testing and filled in a learning style questionnaire. Prior to the test, participants' learning styles were determined by means of Reid's Perceptual learning Style Preference Questionnaire. During the pre- and post-testing, the participants were tested on their knowledge of English vocabulary via the Peabody Picture Vocabulary Test IV. During the training, participants from the experimental group took part in a 40-minute WRTS training whereas participants from the control group received 40-minute writing training. We have found that using WRTS for learning vocabulary is indeed effective. However, significant differences in learning outcomes between visual and non-visual learners were not found in both the control and experimental group. These results suggest that WRTS can be implemented in teaching EFL as an effective training tool for both visual and non-visual learners.
V Této práci se zkoumá problematika studování cizích jazyků pomocí moderních informačních technologií. Zvláštní pozornost je věnovaná především st...
详细信息
V Této práci se zkoumá problematika studování cizích jazyků pomocí moderních informačních technologií. Zvláštní pozornost je věnovaná především studovaní jazyků s použitím korpusu. V práci je popsán postup vývoje rozšíření pro internetový prohlížeč s účelem studování cizích jazyků. Specifikem rozšíření je možnost studenta využívat personalizované příklady použitých slov. Personalizace se opírá o slovní zásobu studenta a preferované sféře použití. Příklady použití sebraný ze zpravodajských portálů.
暂无评论