Lecture Speech Research Articles

Influence of Whiteboard Use on Lecturer Speech and Student Note-taking:

Influence of Whiteboard Use on Lecturer Speech and Student Note-taking:

Kesalahan Berbahasa Dalam Presentansi Dosen Politeknik Negeri Bali

The research entitled Language Errors in Lecturer Presentations at the Bali State Polytechnic aims to determine the level of speech language used in lecturer meetings at the Bali State Polytechnic. This is important because there is a tendency for the pronunciation of words, terms and sentences to be less in accordance with the applicable rules, such as the mention of words, terms and sentences which tend to violate the applicable rules. This study used the observation method, while the technique used was recording and note taking. The target of this research is that the lecturers are proud to use good and correct Indonesian, which in the end the lecturers are able to use spoken Indonesian in meetings or other language activities among fellow lecturers. The results of this study indicate that the use of standard words, pronunciation, and word choice are almost in accordance with the applicable rules. This can be seen in the word choice presentation situation with a score of 88.36 classified as an A. In terms of terms, there is a tendency to use foreign and regional terms which are very proud, namely with a score of 95.48. This value is also classified as very good, with a very satisfactory predicate. In contrast, the use of effective sentences in presentations is not satisfactory. This can be seen from the value of the order of the sentence, which only gets a score of 63.95, which is classified as not good. Thus, the average value in the order of words, terms, and sentences is 82.60. It can be said that the use of the Balinese State Polytechnic lecturer's speech in the presentation is very good.

Speaker stance and engagement across disciplines in Lithuanian university lectures: the case of mes ‘we’ in medicine and business administration

To gain a greater understanding of how speakers construct their disciplinary-situated identities and interact with their addressee(s) in Lithuanian spoken academic discourse, this corpus-based exploratory analysis focuses on the use of mes ‘we’ as a marker of stance and engagement in lecturers’ speech in Lithuanian university lectures on business administration and medicine. The data reveals that the lecturers in business administration used only the inclusive mes ‘we’, which is known to promote student involvement and strengthen lecturer-student rapport. The instructors in medicine frequently employed the exclusive reference to indicate their belonging to professional communities and highlight their level of expertise in the discipline, creating a sense of distance between the lecturer and the student audience.

A Bibliographic Review and Meta-Analysis of the Effect of Teaching Behaviors on Learners’ Performance in Videotaped Lectures

Although the discussion about the influence of the instructor-present videos has become a hot issue in recent years, the potential moderators on the effectiveness of an on-screen instructor have not been thoroughly synthesized. The present review systematically retrieves 47 empirical studies on how the instructors’ behaviors moderated online education quality as measured by learning performance via a bibliographic study using VOSviewer and meta-analysis using Stata/MP 14.0. The bibliographic networks illustrate instructors’ eye gaze, gestures, and facial expressions attract more researchers’ attention. The meta-analysis results further reveal that better learning performance can be realized by integrating the instructor’s gestures, eye guidance, and expressive faces with their speech in video lectures. Future studies can further explore the impact of instructors’ other characteristics on learning perception and visual attention including voice, gender, age, etc. The underlying neural mechanism should also be considered via more objective technologies.

Code-switching for non-native Korean lecturers in teaching Korean as a foreign language for basic and additional level

Although communication is a means of exchanging ideas between people, it can be challenging due to the abundance of synonyms and the evolution of language. The idea is that rather than code-switching, each person needs to have the aptitude and competence to comprehend different kinds of language. This study focuses on the use of code-switching by lecturers in the learning process. The purpose of this study was to find out what types of code-switching were used by lecturers during the learning process, what was the function of using code-switching, and how students perceived the use of code-switching by lecturers in learning activities. This research is based on the awareness that the ability of students to capture learning in the classroom varies. This research is expected to help Korean language learning effectively. The method used in this research was descriptive qualitative, with one lecturer and 71 students as respondents. Data in the form of a transcription of the lecturer’s speeches were obtained from four learning video documentation and questionnaires. Three types of code-switching were found in the lecturer's speech at each meeting, namely the tag-switching type, the intra-sentential-switching, and the intersentential-switching. Then there were five functions realized at each meeting, namely referential functions, emotive functions, conative functions, fatigue functions, and metalinguistic functions. Students’ responses to the lecturer’s code-switching, as evidenced in the questionnaires, were generally positive, especially in terms of understanding the materials.

VLSP 2021 - ASR Challenge for Vietnamese Automatic Speech Recognition

Recently, Vietnamese speech recognition has been attracted by various research groups in both academics and industry. This paper presents a Vietnamese automatic speech recognition challenge for the eighth annual workshop on Vietnamese Language and Speech Processing (VLSP 2021). There are two sub-tasks in the challenge. The first task is ASR-Task1 focusing on a full pipeline development of the ASR model from scratch with both labeled and unlabeled training data provided by the organizer. The second task is ASR-Task2 focusing on spontaneous speech in different real scenarios e.g., meeting conversation, lecture speech. In the ASR-Task2, participants can use all available data sources to develop their models without any limitations. The quality of the models is evaluated by the Syllable Error Rate (SyER) metric.

Shoelace pattern-based speech emotion recognition of the lecturers in distance education: ShoePat23

Background and objectiveWe are living in the pandemic age, and many educational institutions have shifted to a distance education system to ensure learning continuity while at the same time curtailing the spread of the Covid-19 virus. Automated speech emotion classification models can be used to measure the lecturer's performance during the lecture. Material and methodIn this work, we collected a new lecturer's speech dataset to detect three emotions: positive, neutral, and negative. The dataset is divided into segments with a length of five seconds per segment. Each segment has been utilized as an observation and contains 9541 observations. To automatically classify these emotions, a hand-modeled learning approach is presented. This approach has a comprehensive feature extraction method. In the feature extraction, a shoelace-based local feature generator is introduced, called Shoelace Pattern. The suggested feature extractor generates features at a low level. To further improve the feature generation capability of the Shoelace Pattern, tunable q wavelet transform (TQWT) is used to create sub-bands. Shoelace Pattern generates features from raw speech and sub-bands, and the proposed feature extraction method selects the most suitable feature vectors. The top four feature vectors are selected and merged to obtain the final feature vector. By deploying neighborhood component analysis (NCA), we chose the most informative 512 features, and these features are classified using a support vector machine (SVM) classifier using 10-fold cross-validation. ResultsThe proposed learning model based on the shoelace pattern (ShoePat23) attained 94.97% and 96.41% classification accuracies on the collected speech databases consecutively. ConclusionsThe findings demonstrate the success of the ShoePat23 on speech emotion recognition. Moreover, this model has been used in the distance education system to detect the performance of the lecturers.

An Analysis on Speech Styles Used by English Lecturers

The objectives of this research were to (1) find out the types of speech styles used by English lecturers in teaching-learning process and (2) explore the impact of lecturers’ speech style on the students’ attitude towards the lesson. The method of this research was a qualitative descriptive research. There were 2 lecturers and 30 students of 4th semester as the participants of this research. The data were obtained through observation checklist, interview, and video/audio recording. The data from audio recording were transcribed into written form to be analyzed. The result of this research showed that there were 4 speech styles were used by the English lecturer such as formal style, consultative style, casual style, and intimate style. The lecturers used 2 to 4 speech style in teaching-learning process. The impact of lecturer speech style on the students’ attitude towards the lesson, based on the result of the observation and the interview with the students, can be concluded: first, the use of formal speech style did not make the students participate actively and enthusiastically in the lesson even though some students feel that the use of formal style in teaching-learning process could give them more lesson on the language and some feel it is easier to understand the lesson. It is because they felt anxious about the grammatical error that may occur when they speak. Second, the use of consultative style motivated the students to learn more because they felt challenging when the teacher asked them their prior knowledge or opinion regarding the lesson material. The students also participate actively in the class discussion because they think when their lecturers ask them questions; their lecturer is testing their knowledge and ability. It is also stimulated by the belief about effective students is the one who participate actively in the class. Third, the students participate actively and enthusiastically when the lecturer used casual speech style because they do not have to be anxious of the grammar when they speak. The students feel relax and comfortable because the language that is used by the lecturer is the language that they used every day. Fourth, the use of intimate speech style got the students participate actively in the class and got more interested in the lesson. This speech style made the students felt relax and comfortable in the lesson because they felt a close lecturer-students relationship.

Effects of Lecturer Appearance and Speech Rate on Learning Flow and Teaching Presence in Video Learning

Effects of Lecturer Appearance and Speech Rate on Learning Flow and Teaching Presence in Video Learning

Insulin's Discovery: New Insights on Its Hundredth Birthday: From Insulin Action and Clearance to Sweet Networks.

In 2021, the 100th anniversary of the isolation of insulin and the rescue of a child with type 1 diabetes from death will be marked. In this review, we highlight advances since the ingenious work of the four discoverers, Frederick Grant Banting, John James Rickard Macleod, James Bertram Collip and Charles Herbert Best. Macleoad closed his Nobel Lecture speech by raising the question of the mechanism of insulin action in the body. This challenge attracted many investigators, and the question remained unanswered until the third part of the 20th century. We summarize what has been learned, from the discovery of cell surface receptors, insulin action, and clearance, to network and precision medicine.

The impact of semantic annotation techniques on content-based video lecture recommendation

Increasing videos available in educational content repositories makes searching difficult, and recommendation systems have been used to help students and teachers receive a content of interest. Speech is an important carrier of information in video lectures and is used by content-based video recommendation systems. Although automatic speech recognition (ASR) transcripts have been used in modern video recommendation systems, it is not clear how annotation techniques work with noisy text. This article presents an analysis on a set of semantic annotation techniques when applied to text extracted from video lecture speech and their impact on two tasks: annotation and similarity analysis. Experiments show that topic models have good results in this scenario. Besides, a new benchmark for this task has been created and researchers can use it to evaluate new techniques.

Lecturers’ Speech Acts in Arabic Language Learning Interactions at UIN Sulthan Thaha Saifuddin Jambi

This study aims to describe the type and function of lecturers’ speech acts in the interaction of learning Arabic at UIN Sulthan Thaha Saifuddin Jambi. Where in the classroom, the lecturers had important roles in the teaching and learning process. They always use speech as medium to convey ideas to students. Lecturers are also expected to function as active communicators in Arabic lectures. The interaction that is built by lecturers and students is inseparable from the speech act. A qualitative approach was implemented where the data taken from lecturer utterances in Arabic learning interactions were obtained from Arabic lecturers. Data collection was taken by using non-participant observation technique. Mechanical devices and stationery are positioned as data collection aids. Data analysis includes data reduction, data presentation, and data collection/verification results. The finding showed that the lecturers use five types of speech acts in Arabic learning interactions, namely directive, assertive, co missive, expressive, and declarative acts. The finding also showed that there are social and educational functions used by Arabic lecturers. Social functions consist of four categories, namely, competitive functions, convivial functions, collaborative functions, and conflictive functions. While, the educational function of lecturer speech acts consists of nine types, namely, giving a positive impression of the material, directing focus, inclusive learning situations, exclusive learning situations, expressing specific things, encouraging students to build conclusion, controlling the learning situation, and providing motivation to students. The results indicated that the lecturers' speech acts deal with social problems and learning objectives.

RHETORIC STRATEGY IN CLASSROOM DISCUSSION

This article aims to describe the pattern of question strategies and strategies for answering in class discussion activities. The documentation technique is used in collecting data sourced from verbal data, namely the speech of students who present the role as presenters and discussion participants, and lecturers who respond to the results of the discussion. The results of the analysis show that there is a pattern of rhetorical strategies generated in the speech of students in class discussion presentations. The pattern includes question strategies and answer strategies. In the question strategy in class discussion there are four formulas, among others; (1) description + question + marriage - is there in the speech of the audience and questioner ,(2) description + question + explanation is in the lecturer speech, (3) question + description is in the questioner's speech, (4) question + explanation + question is in the speech questioner. Whereas in the strategy of answering in class discussions there are two strategies found, among others; (1) the directanswerstrategy is found in the answering speech and (2) the indirectanswerstrategy is found in the answering speech.

HMM-based phonetic engine for continuous speech of a regional language

A Phonetic Engine (PE), phonetic-level speech recognition system for continuous speech of a regional language named Punjabi, has been proposed in this paper. Punjabi is a highly prosodic language and a very small amount of work has been done in this direction on this language. As a first step towards the development of PE, 25 hrs of data is collected in three different modes, namely, Read Speech, Conversational Speech and Lecture Speech. The 10 hrs of collected data is then manually transcribed using International Phonetic Alphabet (IPA) chart. The architecture of the PE includes three different phases: data preparation, system training and system testing. Initially, the vocabulary of 49 phones is chosen by carefully analyzing the symbol frequency in IPA transcription and data files are prepared to train the system accordingly. The prepared data files and speech files are then used for modeling and feature extraction. In the development of PE, Mel Frequency Cepstral Coefficient (MFCC) is used as a feature extraction technique and Hidden Markov Model (HMM) as a classifier. The PE is developed using HTK Toolkit. The performance of PE is evaluated using three different approaches: (i) by increasing the amount of data from 3 hrs to 5 hrs, (ii) by decreasing the number of symbols from 49 to 29 and (iii) by increasing MFCC dimensions from 12 to 36. An accuracy of 72.3% has been achieved in this work when 5 hrs data with 29 symbols and 12 MFCCs was employed.

MEKANISME EUFEMISME DAN SENSORISASI: KEKERASAN SIMBOLIK DALAM TUTURAN DOSEN

Abstract: Symbolic Violence Representation in the Mechanism of Euphemism and Sensori-zation of Lecturer Speech. This study aims to describe the representation of symbolic violence in the mechanism of euphemism and censorship of lecturers' speeches to students in the class. This study uses a qualitative approach with a type of case study research. Data in the form of sentences in lecturer speech to students. Data collection techniques used are observation, recording, and recording. Bourdieu's social class theory which was constructed in the mechanism of euphemism and censorship became the basis for studying this research data. Symbolic violence is represented in the mechanism of euphemism and the censorship mechanism. The results of the study show that symbolic violence through lecturer speech inspired by institutional authority shapes the behaviors, thoughts, and beliefs that can be accepted in certain social spaces. This is done on the basis of "to educate" students to behave and act better.

Prototype Design of Playback and Search System for Lecture Video Content using Google Cloud API

Recently, it has become possible to easily use video teaching materials in lectures and exercises using high-performance information terminals. In this study, we design and develop playback functions for lecture videos as learning support using a learning management system. The main feature of the proposed system generates keywords from handwritten characters and lecture speech so that learners can search for playback positions related to keywords. Additionally, it becomes possible to create a highly maintainable system in a short development period using speech recognition and still-image recognition functions available on the internet. In this paper, we describe our prototype implementation and present the results of our initial evaluation. The results demonstrate that, regarding the playback of a lecture video, it is possible to use the keyword search function 20–40 min after the lecture video is released. Moreover, we implemented a playback system using a block function on Moodle so that other teaching material in the course could be viewed simultaneously.

Ameliorated language modelling for lecture speech recognition of Indian English

A great amount of research is growing towards the automatic transcription of lectures that consist of numerous information and knowledge that could be helpful to the educational systems and institutes. In large vocabulary speech recognition, language model plays a paramount role in reducing the humongous search space. However, language modelling is very brittle when moving from one domain to another or when moving from read speech to spontaneous speech. Also, lecture speech recognition will have some of the characteristics of spontaneous speech. Hence, it is very challenging to build the language model for this task. In this paper, a judicious approach to adapt the language model in a way where the language model will be in close proximity to the topic spoken in the lecture speech has been depicted. The evaluation of the language model is devised using the proposed approach with the existing language models such as CMU Sphinx, Gigaword and HUB-4. We observed the results analysis that the language models devised from the proposed approach outperform from the existing language models in terms of word error rate, perplexity and out of vocabulary rate. Analysis shows that the presented two-phase approach has resulted in an average decrease of the word error rate to be approximately 14% and the perplexity is decreased by half on average.

Discussion-facilitator: towards enabling students with hearing disabilities to participate in classroom discussions

Class participation plays a vital role in the learning process during classroom instruction. Deaf students often have difficulty participating in class discussions. Several studies have shown that deaf people are better able to interpret speech when they can view the lip movements of a speaker. This paper proposes an assistive device, called the Discussion-Facilitator, which aims to enable deaf students to better participate in classroom discussions. This is done by combining the speech-recognised text of the lecture with a live video stream that is zoomed-in on the lecturer's face. The student is also able to write a text response and play it on loud speakers. Nine deaf students conducted a usability test. The results show that viewing lip movements combined with the speech-recognised text of the lecturer contributed to the understanding of the lecturer's speech, and that our prototype makes the engagement of deaf students in classroom discussion more effective.

Error detection and accuracy estimation in automatic speech recognition using deep bidirectional recurrent neural networks

Recurrent neural networks (RNNs) have recently been applied as the classifiers for sequential labeling problems. In this paper, deep bidirectional RNNs (DBRNNs) are applied to error detection in automatic speech recognition (ASR), which is a sequential labeling problem. We investigate three types of ASR error detection tasks, i.e. confidence estimation, out-of-vocabulary word detection and error type classification. We also estimate ASR accuracy, i.e. percent correct and word accuracy, from the error type classification results. Experimental results using English and Japanese lecture speech corpora show that the DBRNNs greatly outperform conditional random fields (CRFs) and the other NN structures, i.e. deep feedforward NNs (DNNs) and deep unidirectional RNNs (DURNNs). These performance improvements are because the DBRNNs can take the longer bidirectional context of input feature vectors into account and can model highly nonlinear relationships between the input feature vectors and output labels. In detailed analyses, the DBRNNs show a better generalization ability than the CRFs. These results are thanks to the ability of the DBRNNs to represent (embed) the words in a low-dimensional continuous value vector space. In addition, the superiority of the DBRNNs to the DNNs and DURNNs indicates that the average length of the context of the input feature vectors required for ASR error detection is only a few time steps, however, it will change (lengthen) depending on the situation.

Estimating Speech Recognition Accuracy Based on Error Type Classification

Methods for estimating the speech recognition accuracy without using manually transcribed references are beneficial to the research and development of automatic speech recognition technology. This paper proposes recognition accuracy estimation methods based on error type classification (ETC). ETC is an extension of confidence estimation. In ETC, each word in the recognition results (recognized word sequences) for the target speech data is probabilistically classified into three categories: the correct recognition (C), substitution error (S), and insertion error (I). Deletion errors (D) that can occur at interword positions in the recognition results are also probabilistically detected. By summing these CSID probabilities individually, the numbers of CSIDs and, as a result, the two standard recognition accuracy measures, i.e., the percent correct and word accuracy (WAcc), for the speech data can be estimated without using the reference transcriptions. Two recognition accuracy estimation methods based on ETC are proposed. In the first easy-to-use method, ETC is performed by converting the recognition results represented as word confusion networks into word alignment networks (WANs). In the second and more accurate method, the WAN-based ETC results are refined with conditional random fields (CRFs) using various types of additional features extracted for each of the recognized words. Experiments using English and Japanese lecture speech corpora show that the recognition accuracy can be accurately estimated with the CRF-based method. The correlation coefficient and root mean square error between the lecture-level true WAccs calculated using the reference transcriptions and those estimated with the CRF-based method are 0.97 and lower than 2%, respectively. A series of additional experiments and analyses are also conducted to better understand the effectiveness of the CRF-based method.