Real-life Conversations Research Articles

ISETECH-AI-Powered Dental Consultation Chatbot: Enhancing Public Healthcare Accessibility and Awareness

The evolving landscape of healthcare has witnessed remarkable transformations owing to the integration of Artificial Intelligence (AI) and chatbot technology. This paper presents the development of an AI-powered dental consultation chatbot aimed at revolutionizing dental healthcare accessibility and awareness, particularly in the context of Malaysia. The chatbot serves as a valuable resource for dental self-diagnosis, information dissemination, and remote consultations, reducing the need for in-person visits to dental clinics, especially considering the COVID-19 pandemic. The adoption of chatbots in healthcare has gained momentum, driven by their ability to engage in natural language conversations, disseminate reliable information, and provide personalized assistance. Chatbots have emerged as essential tools in public health management, offering services ranging from disease understanding and information provision to counselling and facilitating doctor-patient communication. In the specific domain of dental healthcare, chatbots have the potential to alleviate challenges associated with iatrophobia (fear of disclosing dental problems) and the need for professional dental consultations. The objective of this work is to develop a user-friendly, AI-powered dental consultation chatbot accessible online. This chatbot allows users to engage in real-life conversations with the machine, conducting short self-diagnosis of their oral health conditions. Users can seek information about their dental issues, understand symptoms, and gain insights into their dental health before or after visiting a dentist. Furthermore, the chatbot system presented in this study holds the potential to ease the workload of dental practitioners by offering an alternative method of consultation. Modern digital dental technologies are increasingly focused on diagnosis and classification tasks, but few have explored the realm of dental consultation chatbots for public use. This project bridges this gap and extends the utility of AI-powered chatbots to the field of dental healthcare. This AI-powered dental consultation chatbot not only addresses the evolving healthcare needs of the post-pandemic era but also contributes to advancing dental healthcare accessibility and awareness among the population.

The Future of Artificial Intelligence in English Language Teaching: Pros and Cons of ChatGPT Implementation through a Systematic Review

The integration of artificial intelligence (AI) into language instruction has presented new opportunities, with ChatGPT emerging as a promising tool for interactive and personalized learning. This systematic review examines the effectiveness, advantages, and drawbacks of the ChatGPT in English language teaching (ELT). To achieve this objective, peer-reviewed studies published between 2023 and 2024 were sourced from the Web of Science database and focused on those that explored ChatGPT for language instruction. Using standardized criteria, data extraction covered the study design, participants, interventions, outcomes, and quality assessments. The findings indicate that the ChatGPT significantly improves language teaching by providing personalized feedback, fostering learner autonomy, enhancing student motivation and engagement, and facilitating specific language skills. Moreover, its ability to simulate real-life conversations makes it an effective tool to improve language fluency and comprehension. Nevertheless, challenges have been highlighted, such as overreliance on AI, academic dishonesty, skill deterioration, biases in AI-generated content, and the digital divide affecting access to AI technology. Integrating AI into existing educational frameworks and addressing the need for teacher training have also emerged as significant concerns. The implications are provided based on these findings.

Moving beyond us-versus-them polarization towards constructive conversations

ABSTRACT The global surge of political polarization poses a significant threat to liberal democracies. The prevailing “us-versus-them” mentality prevents leaders from effectively addressing societal issues. While intergroup dialogue shows promise in bridging divisions among diverse identity groups, the dynamics of real-life conversations between individuals with opposing political identities remain underexplored. This study investigates a unique case study of constructive face-to-face interactions among American political elites on a contentious issue. By examining how opposing leaders collaborate towards shared goals, the research identifies communicative actions that can reduce polarization. This study delves into the complex dynamics of polarization, as both an issue-based and identity-based conflict, focusing on interactional framing strategies leaders use to navigate their differences. The findings reveal that political leaders bridge ideological divides by embracing multiple frames, reconnecting conflicting frames, and developing neutral non-political frames. They also foster positive relations, use superordinate identities, and decrease social distance, thereby bridging their identity gap. By fostering convergence rather than accentuating differences, they effectively counteract polarization. Studying rare examples of constructive bipartisan collaboration offers valuable insights into reducing polarization, restoring political trust among the general public, and mitigating broader societal impacts, ultimately strengthening the foundations of a healthy democracy.

The urgency of incorporating real-life conversation into IFL learning materials

While authenticity plays a vital role in foreign language learning, a significant disparity exists between classroom-based learning and the practical use of language in real-world contexts. IFL resources often lack authenticity, failing to mirror the genuine language encounters that learners face beyond the classroom. Therefore, this research explores the motivations, needs, and prior learning experiences of IFL learners to emphasize the importance of integrating real-life conversations into IFL learning materials. This descriptive qualitative research gathered data via surveys and semi-structured interviews with ten foreigners. The findings underscored the varied motivations of participants in learning Indonesian, including personal and familial connections, communication requirements during travel, interests in language, and cultural identity. Participants engaged in instructor-led and self-directed learning as endeavors to boost their proficiency. Intriguingly, there was a discrepancy between their learning experiences and needs. Although the provided IFL materials had successfully fulfilled the learners’ needs in formal settings, they fell short of adequately equipping them for real-world interactions in the intended social settings. Hence, it is suggested that authentic, real-life conversation be incorporated into the IFL materials to meet the diverse practical communication needs of IFL learners. AbstrakKeautentikan memainkan peranan penting dalam pembelajaran bahasa asing, namun pada kenyataannya terdapat kesenjangan yang signifikan antara pembelajaran di kelas dan penggunaan bahasa dalam konteks kehidupan nyata. Sumber belajar Bahasa Indonesia bagi Penutur Asing (BIPA) kerap kali kurang mencerminkan penggunaan bahasa Indonesia sesungguhnya yang dihadapi pemelajar di luar kelas. Oleh karena itu, penelitian ini bertujuan untuk mengeksplorasi motivasi, kebutuhan, dan pengalaman belajar pemelajar untuk menekankan pentingnya pengintegrasian percakapan real-life dalam materi pembelajaran BIPA. Peneliti mengumpulkan data melalui survei dan wawancara semi-terstruktur. Hasil penelitian deskriptif kualitatif ini menyoroti beragam motivasi pemelajar dalam mempelajari bahasa Indonesia, yakni hubungan pribadi dan keluarga, kebutuhan komunikasi saat bepergian, minat terhadap bahasa, serta identitas budaya. Para pemelajar belajar bahasa Indonesia secara formal dengan instruktur dan secara mandiri sebagai usaha untuk meningkatkan kecakapan berbahasa Indonesia mereka. Meskipun begitu, masih ditemukan kesenjangan antara pengalaman belajar dan kebutuhan belajar para pemelajar BIPA. Walaupun materi BIPA yang tersedia sudah mengakomodasi kebutuhan berbahasa pemelajar di konteks formal, materi tersebut belum memadai dalam mempersiapkan pemelajar untuk interaksi autentik dalam konteks sosial yang dibutuhkan. Berdasarkan temuan dalam penelitian ini, disarankan untuk mengintegrasikan percakapan real-life ke dalam materi pembelajaran BIPA guna memenuhi kebutuhan komunikasi praktis pemelajar BIPA yang beragam.Kata Kunci: Materi autentik; materi BIPA; percakapan kehidupan nyata

Formulating parents' feelings: Analyzing parent-nurse conversations in family-integrated neonatal care to develop communication training

ObjectiveThe novel concept of Family-Integrated Care (FICare) requires nurses to be parents' partners in neonatal care. We combined analyses of real-life parent-nurse conversations and interviews to elucidate nurses' role in providing psychosocial support to parents. Findings inform the development of communication training on topicalizing parents' feelings. MethodsConversation analysis of 15 audio-recorded parent-nurse conversations, and thematic analysis of interviews with 2 nurses. ResultsIn parent-nurse conversations, nurses showed a “balancing act” in formulating parents' feelings, revealing the complexities of addressing parents' feelings. Overall, parents confirmed nurses' formulations, but also expanded or modified them, or indicated restricted conversational space. In the interviews, nurses discussed four purposes of conversations with parents, emphasizing elaborating on parents' feelings, while discussing associated challenges. ConclusionOur conversation analysis revealed a continuum of nurses' formulations of parents' feelings, and nurses' reflections illuminated how and when the formulations were used to invite parents' “feelings talk”. InnovationThis study is the first to use conversation analysis to analyze parent-nurse conversations. Additionally, it pioneers combining these analyses with interviews, inviting nurses to reflect on how to incorporate the findings into FICare. This combination strongly informs the development of tailored communication training, drawing from real-life conversations and nurses' articulated needs.

Measuring the Mediated Imagined Interaction Hypothesis: Scale Development

Expanding the Mediated Imagined Interaction Hypothesis, this article contributes the development of a scale to test the theoretical construct. Interweaving imagined interactions and media, the mediated imagined interaction hypothesis examines how media impacts how people imagine their real life conversations and interactions with others. The multiphase scale development and testing process for the Mediated Imagined Interaction Scale are presented in this article. Further explorations for application of this scale are discussed.

The Representation of Pragmatic Knowledge in Moroccan High School English Language Teaching (ELT) Textbooks

Studies in the domain of English language teaching reveal a significant disparity between pragmatic findings and the approaches employed in English language instruction (Ishihara & Cohen, 2010). Investigations into the effectiveness of textbooks for teaching communicative skills, mirroring real-life conversations, show that English Language Teaching (ELT) textbooks often lack clear and thorough explanations of English conversational dynamics (Berry, 2000; Burns, 1998; Cane, 1998; Grant & Starks, 2001). According to the available literature on EFL/ESL textbooks, speech acts, and other pragmatic elements are largely neglected in both quality and quantity (e.g., Vellenga, 2004; Salazar Campillo, 2007; Neddar, 2010). Therefore, this study examines how textbooks cultivate pragmatic competence and evaluates the pragmatic content in EFL textbooks used in Moroccan high schools from 2006-2007. It aims to assess the depth and quality of pragmatic information covered in these textbooks. In addition to concentrating on metapragmatic details, the study discusses register, illocutionary force, politeness, appropriateness, and usage, as well as how textbook authors present speech acts. However, the study findings reveal that the selected textbooks provide insufficient pragmatic information. This underscores the necessity of supplementing EFL textbooks with more authentic materials and ample meta-pragmatic information. It is also essential to consider pragmatics during the textbook design process and enhance teachers' abilities to incorporate them into their classrooms.

"Dentro. Una storia vera, se volete / Dentro (Inside). A True Story, If You Think So" by Giuliana Musso

Giuliana Musso’s fearless play about the structures and cultures that hide sexual abuse is a Pirandellian spiral of truths that elude proof, presented through a theatrical framing that confounds reality with art. The most recent play of nine that she has written and performs in her preferred “theater of inquiry” style, Dentro (Inside), dramatizes Musso’s real-life encounter with an acquaintance who shared an impossible story of incest. In the opening minutes, Roberta discloses the horrifying experience of slowly learning that her ex-husband molested their daughter Chiara for years when they still lived together. Over a series of “chapters,” or scenes, it becomes apparent how every system in place to help vulnerable members of society—from the hospitals to private psychologists, the police and law enforcement, eventually the lawyers, the whole legal system, and even language itself—was unable to support Roberta and her daughter. To the contrary, given that the case was dismissed, they have gag orders; they cannot even speak of what happened so as not to slander the reputation of the ex-husband/ father. In a last-ditch effort, Roberta turns to Giuliana, who she knows as a celebrated actor-author, to devise a play that speaks to these issues. The actual play is ultimately a dramatized account of their real-life conversations.

EMBRACE (Empowering Medical students' skills in BReaking bAd news with Compassion and Empathy)module improves the skills of undergraduate medical students in effectively breaking the bad news: a case-control study.

Effective communication skills are pivotal in health care, particularly when conveying distressing information to patients and their families. However, medical education still lacks the adoption of a universal model that can be incorporated into the curricula to train and assess students in effectively communicating with patients. This study aims to assess the impact of training undergraduate medical students to deliver bad news effectively using theEmpowering Medical students' skills in BReaking bAd news with Compassion and Empathy (EMBRACE) module. This randomized case-control study involved medical students from the first, second, and third professional years (study group, n = 75; control group, n = 75). For the study group, the EMBRACE modules were distributed. Then, a 1-hour training session on effectively delivering bad news was followed by a multiple-choice question test and objective structured clinical examination with response, interpretation, and communication skills stations. Participants' feedback was obtained on a five-point Likert scale. There was a highly significant improvement in knowledge and skills among the study group compared to controls with a P value less than 0.0001. Of the participants, 98.76% perceived that the training equipped them with practical skills, and 98.77% felt that the facilitator had demonstrated the steps of delivering bad news clearly and effectively. Only 4.44% of participants were confident in effectively interacting with patients before the session, and an overwhelming 81.11% gained confidence in their communication skills after the training. With demonstrated significant improvement in knowledge and skills, this study supports the adoption of EMBRACE modules in undergraduate medical education, ultimately improving patient experiences, doctor-patient relationships, and health outcomes.NEW & NOTEWORTHY TheEmpowering Medical students' skills in BReaking bAd news with Compassion and Empathy (EMBRACE) module is noteworthy for its holistic approach to training medical students in the delicate art of delivering distressing news to patients. It not only incorporates the evidence-based setting, perception, invitation, knowledge, emotions, and strategy (SPIKES) method but also distinguishes itself by providing real-life conversation examples and self-assessment cases, which make the training highly relatable and practical for students to actively engage in their learning and personal development.

Case study regarding adult students’ learning of world languages and cultures in a unique voluntary and learner-centred nonformal setting

The purpose of this qualitative study was to examine a group of adult students’ learning of world languages and cultural practices by incorporating common literacy practices (listening, reading, speaking, writing) in a unique, voluntary, learner-centred, nonformal setting (e.g. World Language and Culture Program [WLCP]) at a large western land-grant university in the United States during the 2019 spring semester. The primary objectives of this study are to provide insight into what and how adults learned in this unique environment. Drawing from Simons, the study was framed as a qualitative descriptive case study. Data sources included interviews, observations, field notes, and artefacts. Data analysis drew on the Deep Approach conceptual lens and the Interpret Analyse Present Interact (IAPI) curriculum model to understand adult students’ learning in this WLCP. Findings showed that adults learned real-life conversation skills and cultural practices (geography, traditions, religious practices) by incorporating common literacy practices (listening, reading, speaking, writing) from volunteer facilitators who were natives of the particular world language and culture. The participants learned these world language conversation skills and cultural practices in a variety of learner-centred ways: (a) practice with heritage speakers of the language, (b) meaningful interactions with peers, (c) contextualised simulations, (d) spontaneous exposure, and (e) instructional scaffolding.

The Effect of GPT Chat on Indonesian and Italian National Security

The features contained in ChatGPT are centered on providing conversations. One of ChatGPT's capabilities is that it can imitate real-life conversations because its learning is further supervised and uses a perfect language model, so its coverage differs from other smart technologies. ChatGPT has a way of expanding its functions and features, namely with plugins that can provide services to users so that this system can maximize applications. With other plugins, ChatGPT can access and control these third-party services. ChatGPT's biggest weakness is that it sometimes produces text that when read sounds correct but is incorrect information. ChatGPT also does not provide clear references, resulting in copyright problems. Therefore, ChatGPT has been banned by several countries due to copyright issues. The Italian government immediately blocked ChatGPT because there were no special regulations regarding this application. Meanwhile, the Indonesian government has not yet responded to the ChatGPT problem. The two countries have different views regarding ChatGPT, especially regarding national security.

Polar Influence of Code-Switching on Oral Communication Skills: The Case of Philippine Cebuano-Visayan Bilinguals

Linguists are explicitly divided on code switching's (CS hereinafter) effect on the oral communication skills of English language learners. Some studies asserted that CS is beneficial (positive) while others criticized its negative impact on the target language (TL hereinafter) proficiency in communication skills. This study examined the influence of CS strategy on the oral communication skills of bilingual Cebuano-Visayan Senior High School students of the University of Cebu - Banilad, Cebu City, Philippines. This serves to bridge the ongoing debate and as a recent contributing perspective on the CS phenomenon, especially the English and Cebuano-Visayan language pair not yet explored in a conversational analysis in a new dialectal context (locale). Employing a case study design with 9 participants for data saturation, real-life conversations of students and teachers in English classrooms were audio-recorded for one month, and triangulated with interviews and observations. Thematic and conversational analysis then ensued which yielded the finding that CS adversely affects the students' oral communication skills in the areas of morphology, phonology, and syntax but positively facilitates the strategic competence of the bilingual oral communicators. CS, therefore, has a polar influence (both positive and negative) on the student’s oral communication skills.

Use of an online gaming tool, the Veterinary DialogueTrainer, for teaching clinical communication skills to bovine veterinary practitioners.

Effective clinical communication is essential for bovine veterinary practitioners to establish and maintain positive client relationships. When applied properly in herd health consultancy, it increases client satisfaction and adherence to veterinary advice, and improves patient health and welfare. Although communication skills are often taught by providing feedback on simulated conversations, this has limitations, including time constraints, subjective assessments, and cost. The Veterinary DialogueTrainer (VDT) is an online serious game platform using 'digital role-play' with avatars, recently developed to enhance and assess learning outcomes, improve use of learned skills, and increase cost-effectiveness of communication training. The objective was to evaluate its suitability and applicability. Finnish (n = 24) and Swedish (n = 21) bovine veterinarians participated in communication training using VDT for training and assessment. Participants completed the provided scenario at least twice. After playing a bovine health communication simulation, participants received their scores and feedback on selected conversation options. VDT scores measured multiple aspects of communication on a 0-100% scale, based on motivational interviewing methodology and Calgary-Cambridge guidelines. Mean (±SD) number of attempts participants played the scenario was 4.1 (±2.6, Finland) and 3.9 (±1.3, Sweden), with highest total score reached after a mean of 3.5 (±2.1, Finland) and 3.1 (±1.1, Sweden) attempts. On the first attempt, 39 participants (87%) scored <50% of the highest possible score, whereas most (n = 34, 76%) achieved a higher score on the second attempt. Mean total score increased from 15 (±14) to 77% (±33) for Finish participants and from 40 (±22.0) to 87% (±19.4) for Swedish participants. The majority (n = 33, 73%) of participants reached a score >80% after 4.0 (±1.6, Finland) or 2.8 (±1.0, Sweden) attempts. Net Promoter Score of the training was +89 (Finland), +88 (Finland) and + 83 (Sweden) on a scale from -100 to +100, indicating that most participants were very likely to recommend the training to other veterinarians. Use of VDT increased communication skills scores but whether it will improve communication skills in practice requires further study. We concluded there is a likely benefit of using the VDT in teaching and monitoring veterinary communication competencies and preparing for offline role-plays and real-life conversations in veterinary practice.

General practice undergraduate and vocational training: ambulatory teaching and trainers’ curriculum and remuneration – a cross-sectional study among 30 member countries of WONCA Europe

BackgroundAfter a long phase without any propositions for real ambulatory training inside general practitioners’ offices, general practice (GP) vocational training has begun to appear progressively and has been integrated into undergraduate medical programmes. The aim of this study was to provide an overview of GP vocational training and GP trainers in member countries of the World Organization of National Colleges, Academies and Academic Associations of General Practitioners/Family Physicians (WONCA) Europe.MethodWe carried out this cross-sectional study between September 2018 and March 2020. The participants responded to a questionnaire in real-life conversations, video conferences or e-mail exchanges. The respondents included GP trainers, teachers and general practitioners involved in the GP curriculum recruited during European GP congresses.ResultsRepresentatives from 30 out of 45 WONCA Europe member countries responded to the questionnaire. Based on their responses, there is a well-established period for GP internships in undergraduate medical programmes, but with varying lengths. The programmes for some countries offer an internship after students graduate from medical school but before GP specialisation to ensure the career choice of the trainees. After specialisation, private practice GP internships are offered; however, in-hospital GP internships are more common. GP trainees no longer have a passive role during their internships. GP trainers are selected based on specific criteria and in countries, they have to follow some teacher training programmes. In addition to income from medical appointments carried out by GP trainees, GP trainers from some countries receive additional remuneration from various organisations.ConclusionThis study collected information on how undergraduate and postgraduate medical students are exposed to GP, how GP training is organised and the actual status of GP trainers among WONCA Europe member countries. Our exploration of GP training provides an update of the data collected by Isabel Santos and Vitor Ramos in the 1990s and describes some specificities that can inspire other organisations to prepare young, highly qualified general practitioners.

The Study of Opening and Closing of Conversations in Oral English Textbooks

Language communication ability refers to the ability to use language in society. The acquisition of this ability by second language learners mostly relies on textbook dialogues adapted from real-life conversations. Therefore, whether scripted textbook conversations can help improve second language learners’ language communication ability should be taken seriously by the academic circle. This paper has analyzed all the conversational texts in the book Functioning in an Intercultural Community, using book1 and book2 as corpus and finds that there are deficiencies in the opening and closing of the conversation in the textbook. This study suggests that textbook compilers should improve the completeness, diversity, and colloquialism of the beginning and end of conversations, and teachers appropriately extend the form of the opening and closing of the conversation during the teaching process.

Argumentation in end-of-life conversations with families in Dutch intensive care units: a qualitative observational study

PurposeIn intensive care units (ICUs), decisions about the continuation or discontinuation of life-sustaining treatment (LST) are made on a daily basis. Professional guidelines recommend an open exchange of standpoints and underlying arguments between doctors and families to arrive at the most appropriate decision. Yet, it is still largely unknown how doctors and families argue in real-life conversations. This study aimed to (1) identify which arguments doctors and families use in support of standpoints to continue or discontinue LST, (2) investigate how doctors and families structure their arguments, and (3) explore how their argumentative practices unfold during conversations.MethodA qualitative inductive thematic analysis of 101 audio-recorded conversations between doctors and families.ResultsSeventy-one doctors and the families of 36 patients from the neonatal, pediatric, and adult ICU (respectively, N-ICU, P-ICU, and A-ICU) of a large university-based hospital participated. In almost all conversations, doctors were the first to argue and families followed, thereby either countering the doctor’s line of argumentation or substantiating it. Arguments put forward by doctors and families fell under one of ten main types. The types of arguments presented by families largely overlapped with those presented by doctors. A real exchange of arguments occurred in a minority of conversations and was generally quite brief in the sense that not all possible arguments were presented and then discussed together.ConclusionThis study offers a detailed insight in the argumentation practices of doctors and families, which can help doctors to have a sharper eye for the arguments put forward by doctors and families and to offer room for true deliberation.

Do Visual Cues Aid Comprehension of a Dialogue?

Do Visual Cues Aid Comprehension of a Dialogue?

Functional neuroanatomy of lexical access in contextually and visually guided spoken word production.

Lexical access is commonly studied using bare picture naming, which is visually guided, but in real-life conversation, lexical access is more commonly contextually guided. In this fMRI study, we examined the underlying functional neuroanatomy of contextually and visually guided lexical access, and its consistency across sessions. We employed a context-driven picture naming task with fifteen healthy speakers reading incomplete sentences (word-by-word) and subsequently naming the picture depicting the final word. Sentences provided either a constrained or unconstrained lead-in setting for the picture to be named, thereby approximating lexical access in natural language use. The picture name could be planned either through sentence context (constrained) or picture appearance (unconstrained). This procedure was repeated in an equivalent second session two to four weeks later with the same sample to test for test-retest consistency. Picture naming times showed a strong context effect, confirming that constrained sentences speed up production of the final word depicted as an image. fMRI results showed that the areas common to contextually and visually guided lexical access were left fusiform and left inferior frontal gyrus (both consistently active across-sessions), and middle temporal gyrus. However, non-overlapping patterns were also found, notably in the left temporal and parietal cortices, suggesting a different neural circuit for contextually versus visually guided lexical access.

Implicature in John Green’s The Fault Of Our Stars

Discourses on revealing meanings in utterances are widely discussed in some fields of linguistics, semantics, presupposition, and implicature. This study is an example of how implicature is employed in a novel, which is a depiction of real life conversation. The aims of this study are to reveal the use of implicature in a novel by using Grice’s cooperative principle framework and to entangle the implied meanings in the conversations in terms of their maxim arrangement. Qualitative research design was used in this study. The result of the study shows that there are two kinds of implicature, namely generalized and particularized conversational implicature. It is also found that there are five patterns of maxims organization in generalized conversational implicatures and four patterns of maxims organization in particularized conversational implicatures.

Emotional conversation generation with heterogeneous graph neural network

The successful emotional conversation system depends on sufficient perception and appropriate expression of emotions. In a real-life conversation, humans firstly instinctively perceive emotions from multi-source information, including the emotion flow hidden in dialogue history, facial expressions, audio, and personalities of speakers. Then, they convey suitable emotions according to their personalities, but these multiple types of information are insufficiently exploited in emotional conversation fields. To address this issue, in this paper, we propose a heterogeneous graph-based model for emotional conversation generation. Firstly, we design a Heterogeneous Graph-Based Encoder to represent the conversation content (i.e., the dialogue history, its emotion flow, facial expressions, audio, and speakers' personalities) with a heterogeneous graph neural network, and then predict suitable emotions for feedback. Secondly, we employ an Emotion-Personality-Aware Decoder to generate a response relevant to the conversation context as well as with appropriate emotions, through taking the encoded graph representations, the predicted emotions by the encoder and the personality of the current speaker as inputs. Experiments on both automatic and human evaluation show that our method can effectively perceive emotions from multi-source knowledge and generate a satisfactory response. Furthermore, based on the up-to-date text generator BART, our model still can achieve consistent improvement, which significantly outperforms some existing state-of-the-art models.