The enhanced visualization and interaction with information in collaborative VR environments enabled by chatbots is currently rather limited. To fill this gap and create a concrete application that combines spatial and virtual concepts of hypertext systems based on the use of LLMs, we present VR-ParlExplorer as a system for virtualizing plenary debates that allows users to interact with virtual members of parliament through chatbots. VR-ParlExplorer is implemented as a Plugin for Va.Si.Li-Lab to enable immersion in the dynamics of communication in parliamentary debates. The paper describes the functionality of VR-ParlExplorer and discusses specifics of the use case it addresses.

CCS Concepts: • Information systems → Collaborative and social computing systems and tools; Multimedia information systems.

Keywords: Hypertext, Parliamentary Speeches, VR-Interaction, Chatbot, RAG

Visualizing the interaction with information units can be a complex task with many different aspects. In particular, the visualization, presentation, and perception of multimodal information is becoming increasingly difficult as interaction capabilities become more sophisticated. The same applies to hypertext systems 12, for which a large number of different applications have been created in the past (see 52 for an overview). In this context, spatial 53, 55, architectural 30, adaptive 7, narrative 6, 32, argumentative 58, taxonomic 42, sculptural 26, virtual 57, and rhetorical hypertext 45 are of relevance, which also address issues of information presentation and interaction in their respective domains. This is also related to an ongoing discussion in the context of hypertext systems in combination with virtual (VR 31, 37), augmented (AR 13, 61) and mixed reality (XR 23, 50), with respect to proportionality and utility 9. In these domains, the concepts of immersion 51, collaboration 56, and interaction 11 with objects and other users are just as important as new and innovative approaches to visualizing data 38, 39.

Following this scenario, and to address different use cases of virtual collaborative hypertexts, a new VR framework called Va.Si.Li-Lab 40 has been developed that addresses a wide range of application areas. In Va.Si.Li-Lab, implemented in Unity, multiple users can communicate with each other in VR and interact collaboratively with virtual objects by means of Ubiq 20. Equipped with a multimodal database, the representation of agent roles using different avatar models (Unity 36, Ubiq, OpenAI 48 and metaverse avatars 16) and multimodal information system functionalities, Va.Si.Li-Lab allows the integration of different Plugins to map the dynamics of face-to-face communication in groups of interlocutors 4, 40. Due to this extensible platform, different application scenarios can be mapped by Va.Si.Li-Lab. In this paper, we add a further scenario: VR-based interaction in the context of virtualized representations of plenary speeches and debates. This is done by the so-called VR-ParlExplorer. VR-ParlExplorer enables automatic processing, presentation and interaction with plenary debates and their virtualization in VR, while extending the functionality of Va.Si.Li-Lab by means of Natural Language Processing (NLP) and Large Language Models (LLM). All speeches, organized in minutes and grouped by agenda items, can be replayed individually using an avatar of the speaker and text-to-speech (TTS) 49 software. Users of VR-ParlExplorer can interact with the speaker avatar using a language model that has been fine-tuned to the speeches of Members of Parliament (MP). Interactions are realized using TTS and speech-to-text 21 (STT) models. To this end, Va.Si.Li-Lab includes a module for DUUI 33 a powerful framework for distributed and scaled NLP. With VR-ParlExplorer, we make the dynamics of parliamentary work visible and accessible in an interactive way that goes beyond the possibilities of a virtual tour of the respective parliament building 46.

Although Va.Si.Li-Lab already includes a multimodal data model, allows collaborative interaction with objects, face-to-face communication while mapping different scenarios using a dynamic event system and different avatars, two crucial features are missing to realize VR-ParlExplorer:

(I) Repurposing of NLP and related methods: Unstructured or semi-structured data cannot be used directly, but must be pre-processed by NLP. This pre-processing should be flexible to ensure that methods and models can be reused for different languages in a structured way, which requires an underlying annotation schema. Processing should also be distributed across a cluster while remaining available locally so that large corpora can be processed, while updates can be handled without additional overhead. This includes TTS and STT capabilities, along with the processing of multimedia information units (video, images, and audio), and their segmentation and conversion. VR-ParlExplorer is equipped in all these respects to provide the required functionality.

(II) LLMs: The use of LLMs is a key feature for a wide range of applications 15, the proper use of which depends on the feature (I). The use of LLMs and their internal optimization open up a wide range of applications, including with regard to ensuring data security. Given the scenario of immersion in parliamentary debates, LLMs enable (1) the enrichment of conversation content with additional information, (2) the transcription of spoken words, (3) the verbalization of transcribed text, and (4) interaction with the resulting data sets using Retrieval Augmented Generation (RAG).

Since there is no VR application that visualizes plenary debates, a direct comparison with VR-ParlExplorer is not possible. Although there are virtual reconstruction projects that explore parliaments through virtual or guided tours (e.g. 47), these projects do not provide a basis for comparison. Since a direct comparison of the tools is insufficient and given Va.Si.Li-Lab's functionality 4, the selection of competitors focuses on tools that visualize information, automatically pre-process it using NLP, and allow for interacting with chatbots in VR. The latter condition is crucial, which means that projects like 22, 54, 62 will not be reviewed here.

Table 1: Comparing VR frameworks according to the features of Section 2. Legend: satisfied (●), partially satisfied (◐), not satisfied (○).

The first system 34 explores the integration of GPT 3.5 in VR for training autistic employees using avatars from Unity and the Lypsync feature from Oculus. Although NLP is not provided, the system supports TTS and TSS (I), as well as interaction with avatars through an LLM (II), but without fine-tuning. Another system is Intention-Aware Human-AI 27, which enables intention-aware cooperation with an LLM (II) that can assist users in performing their tasks in VR. NLP is not supported (I), but speech recognition is employed. In contrast, iVenue 24, which supports various devices including VR, allows virtual tours of a university based on information visualization and navigation, and integrates a chatbot, but without using an LLM (II) or NLP (I). A VR application for Managing School Stress 35 among students uses group counseling to incorporate a chatbot, but without LLM support (II) and without the availability of a flexible pre-processing system, although TTS and STT are available (I). There is also the AI Agents in Social Virtual Reality 63 (AI-Social VR) project, in which individuals can interact with avatars based on VRChat, which generate responses using GPT-4 (II) while using TTS and STT (I), but without any further NLP. Another tool that uses VRChat is Safe Guard 64, an LLM chatbot (II) based on GPT-3.5 and a convolutional neural network to detect hate speech, and SST and TTS (I) to capture and respond to speech. VRChat is also used by ELLMA-T 44, which has been implemented as a VR learning environment for learning English by simulating everyday situations. It uses GPT4 for the virtual avatar (II) and TTS and STT methods (I) from OpenAI 43. Scottish Bonspiel VR 8 enables the preservation and enhancement of immaterial cultural heritage in VR using GPT-4 (II). For this purpose, STT and TTS are available, but no additional NLP (I). VR-Learning 29 present a VR system that implements LLM and RAG (II) in the context of learning tasks, where texts are pre-processed (I) to prepare RAG, but without TTS and STT functionality. Finally, VR-TextEntry 10 is a system that facilitates text production using an LLM-based virtual keyboard (II), but without any other NLP (I).

Regarding the use of NLP and LLMs in VR, most of these systems do not meet the requirements of Section 2 – see Table 1. Their usability is therefore limited and requires further development as implemented by Va.Si.Li-Lab and realized in VR-ParlExplorer, which is described in the next section.

VR-ParlExplorer has been integrated as a Plugin for Va.Si.Li-Lab to profit from this infrastructure. As a result, multiple users can enter the system using different interfaces (VR glasses, desktop, browser) and find themselves in a virtual room (Room) – mapping a parliament, where they have to select a Scene to manifest a session. VR-ParlExplorer facilitates the mapping and rendering of plenary items (e.g., Protocols, Agenda Items, Speechs, Comments, the associated Speakers, Factions etc.) needed to model plenary debates. Using this data structure, debates from different parliaments can be integrated. This is facilitated by the interface-based implementation of the import method within the Java backend; further pre-processing may be necessary depending on the data source. By mapping to these data structures, as shown in Figure 2, an MP's speech can be retrieved within a protocol based on an agenda item, so that it is delivered by the speaker using TTS (see Figure 3). Speeches are often accompanied by interruptions and comments. This additional information is taken into account insofar as i) the comments are faded in and synthesized using TTS from the direction of the MP speaking, and ii) applause or boos are reproduced from the corresponding direction of the plenary hall. VR-ParlExplorer covers a wide range of parliaments so that one can follow a virtual debate remotely.

4.1 Generation of the Environment

VR-ParlExplorer allows a generic representation of parliaments, where each parliament must be mapped individually in terms of visualization and arrangement of its members. This includes the layout of the plenary chamber (Figure 1), the arrangement of its members (MP) in factions and the visualization of the MP's themselves. VR-ParlExplorer currently maps the 20th German Bundestag (2021–2025). To get a realistic representation and a high level of immersion in a plenary chamber, some specific settings are required, but there are also possibilities for automatic mapping if this information is available. If this information is provided, the members of parliament are grouped according to their political factions. VR-ParlExplorer follows five principles that are applied to the automatic placement of MPs to provide a generic representation:

Figure 1: The virtualization of the 20th German Bundestag, viewed from a bird's eye perspective.

MP's faction sit separately from other factions.

Factions that belong to a political movement are seated either on the left or the right.

Seats are generated in rounded rows.

Factions are separated from each other by aisles leading to the speaker's podium.

MP's can sit wherever they wish as long as they remain in the area designated for their faction.

There is one exception: if a faction or group has only a few MPs, they can sit next to each other with no aisle between them. While there is no limit to the number of MPs who can sit in a row, a conical shape is common. Most political factions have at least one seat in the first row. The row structure can be generated automatically by approximating it based on the given knowledge, the number of representatives per faction, and the maximum number of rows. VR-ParlExplorer allows for making distance-related changes to rows and seats, such as adjusting the distance between rows and the distance between MP seats. To create a more precise structure and represent parties sitting next to each other, one can specify a starting position for each party along with an integer array, where the index represents the row and the value represents the number of seats that party has in that row. Each seat contains a colored table representing a party and an object symbolizing an MP from that party, which can be customized. Regardless of the automatic mapping, any parliament can be loaded as long as a VR-ParlExplorer-compliant JSON array containing the necessary information is provided. MPs' portraits can be automatically segmented and projected onto their virtual representatives, making them recognizable and facilitating interaction and communication.

4.2 Corpus

4.3 Natural Language Processing

To visualize the speeches of MPs and use them for interaction via a chatbot, NLP is required. For this purpose, VR-ParlExplorer uses DUUI, which uses distributed, clustered, and schema-based NLP methods using microservices such as Docker and Kubernetes 3. DUUI uses the Apache framework Unstructured Information Management Applications (UIMA). With UIMA, multimodal information can be annotated and made accessible using stand-off annotations and various analysis engines executed by DUUI inside Docker containers to enrich speech documents with annotations. As shown in Figure 4, the plenary protocols are imported into DUUI using a reader and then processed through a pipeline consisting of various components before being stored in a MongoDB. After being mapped to the data structure, the speeches pass through the pipeline to be annotated with spaCy regarding sentence-level sentiments and sentence embeddings for serialization in the database.

Figure 4: NLP of parliamentary protocols using DUUI.

4.4 Chatbot

Each MP is represented as a stand-alone chatbot that the Va.Si.Li-Lab user can interact with. To this end, communication revolves around the RAG architecture design pattern 25 shown in Figure 5. For this purpose, all speeches of the 20th legislative period were collected and converted into embeddings with Jina Embedding V3 59 (Figure 4). Then, when communication with a virtual representative is initiated by looking at them and pressing the appropriate push-to-talk button, VR-ParlExplorer selects the data of the current parliamentary session and converts the user's auditory input into a transcription using WhisperX 5 and then into embeddings. These embeddings were then compared with the existing embeddings of the MP's pre-processed speeches up to the currently selected session, using vector similarity to obtain analogous topics that the MP might have addressed in a speech. These results are sent to an LLM to generate a response using deepseek-r1 14. Next, to create conversations with an MP over multiple messages, both input and response texts are stored inside the virtual MP's object so that they can be passed along with the new input to the next LLM requests. Finally, the result is converted back into an audio stream using Thorsten-Voice 41, whereby each MP's voice has a slightly different pitch, resulting in a more dynamic atmosphere.

Figure 5: VR-ParlExplorer: The RAG process using DUUI.

Based on the prototype resulting from VR-ParlExplorer and the associated implementation of new features within Va.Si.Li-Lab, several developments are on the horizon. Importing and providing access to data across parliaments and legislative periods is a worthwhile next step, especially given the volume of plenary corpora available (e.g. 17, 18). Since the database used to generate the chatbots consists only of plenary speeches from the current legislative period, the next step is to include historical speeches and integrate external sources such as speeches outside of plenary debates and written statements from social media. In addition, existing data sets can be augmented not only with one-dimensional information units, but with multimodal sources, e.g., video and audio. Preprocessing is required in both cases to handle large volumes of data, taking into account models and methods to be incorporated into DUUI. Finally, the design of the MPs and the plenary chamber is not yet finalized; so further design efforts will be required, which will be related to immersion in the given application context.

We presented VR-ParlExplorer, a spatial hypertext system for generating virtual plenary chambers and mapping debates based on the speeches of MPs. VR-ParlExplorer exists as a Plugin for Va.Si.Li-Lab with the goal of reusing this framework's functionalities, such as its multimodal data model, collaborative use of objects and information, and interaction and communication between users. For the realization of VR-ParlExplorer, two yet not fully implemented features were integrated into Va.Si.Li-Lab to enable NLP of unstructured information and the use of virtual avatars based on chatbots. Plenary debates can be partially recreated in VR using VR-ParlExplorer and interaction with a chatbot for each member of parliament has been enabled. VR-ParlExplorer is released on GitHub under the AGPL license to allow flexible development and reuse of this application using Va.Si.Li-Lab. A final note: although VR-ParlExplorer is a VR system that makes speeches immersive, we promote it as a hypertext system. The reason lies less in retrospective analogies with classic hypertext functionalities than in the prospective outlook of text archives as increasingly virtual spaces that can be experienced immersively. In our case, archived speeches become accessible as relivable hypertext spaces thanks to the avataric and architectural qualities of our system.

We gratefully acknowledge the financial support provided by the German Research Foundation (DFG) for the project "New Data Spaces for the Social Sciences" (SPP 2431) – "ENTAILab – Research Infrastructure and Innovation Lab" (539634240)

Giuseppe Abrami, Mevlüt Bagci, Leon Hammerla, and Alexander Mehler. 2022. German Parliamentary Corpus (GerParCor). In Proceedings of the Language Resources and Evaluation Conference, Nicoletta Calzolari, Frédéric Béchet, Philippe Blache, Khalid Choukri, Christopher Cieri, Thierry Declerck, Sara Goggi, Hitoshi Isahara, Bente Maegaard, Joseph Mariani, Hélène Mazo, Jan Odijk, and Stelios Piperidis (Eds.). European Language Resources Association, Marseille, France, 1900–1906. https://aclanthology.org/2022.lrec-1.202

Giuseppe Abrami, Mevlüt Bagci, and Alexander Mehler. 2024. German Parliamentary Corpus (GerParCor) Reloaded. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, and Nianwen Xue (Eds.). ELRA and ICCL, Torino, Italy, 7707–7716. https://aclanthology.org/2024.lrec-main.681

Giuseppe Abrami, Markos Genios, Filip Fitzermann, Daniel Baumartz, and Alexander Mehler. 2025. Docker Unified UIMA Interface: New perspectives for NLP on big data. SoftwareX 29 (2025), 102033. doi:10.1016/j.softx.2024.102033

Giuseppe Abrami, Alexander Mehler, Mevlüt Bagci, Patrick Schrottenbacher, Alexander Henlein, Christian Spiekermann, Juliane Engel, and Jakob Schreiber. 2023. Va.Si.Li-Lab as a Collaborative Multi-User Annotation Tool in Virtual Reality and Its Potential Fields of Application. In Proceedings of the 34th ACM Conference on Hypertext and Social Media (Rome, Italy) (HT '23). Association for Computing Machinery, New York, NY, USA, Article 22, 9 pages. doi:10.1145/3603163.3609076

Max Bain, Jaesung Huh, Tengda Han, and Andrew Zisserman. 2023. WhisperX: Time-Accurate Speech Transcription of Long-Form Audio. In INTERSPEECH 2023 (Dublin, Ireland) (Interspeech 2023). ISCA, 4489–4493. doi:10.21437/interspeech.2023-78

Mark Bernstein. 1999. Structural Patterns and Hypertext Rhetoric. Comput. Surveys 31, 4 (dec 1999), 19–es. doi:10.1145/345966.346011

Peter Brusilovsky. 2001. Adaptive Hypermedia. User Modeling and User-Adapted Interaction 11, 1 (2001), 87–110. doi:10.1023/A:1011143116306

Ka Hei Carrie Lau, Efe Bozkir, Hong Gao, and Enkelejda Kasneci. 2024. Evaluating Usability and Engagement of Large Language Models in Virtual Reality for Traditional Scottish Curling. arXiv:2408.09285 [cs.HC] https://arxiv.org/abs/2408.09285

Chu-Yang Chang, Hsu-Chan Kuo, and Zhengyi Du. 2023. The role of digital literacy in augmented, virtual, and mixed reality in popular science education: a review study and an educational framework development. Virtual Reality 27, 3 (June 2023), 2461–2479. doi:10.1007/s10055-023-00817-9

Liuqing Chen, Yu Cai, Ruyue Wang, Shixian Ding, Yilin Tang, Preben Hansen, and Lingyun Sun. 2024. Supporting Text Entry in Virtual Reality with Large Language Models. In 2024 IEEE Conference Virtual Reality and 3D User Interfaces (VR). IEEE, 524–534. doi:10.1109/vr58804.2024.00073

Leonard Chen, David Goedicke, and Wendy Ju. 2024. Extending ReRun, Annotating Interactions in VR. In Companion of the 2024 ACM/IEEE International Conference on Human-Robot Interaction (Boulder, CO, USA) (HRI '24). Association for Computing Machinery, New York, NY, USA, 1281–1283. doi:10.1145/3610978.3641119

J. Conklin. 1987. Hypertext: An Introduction and Survey. Computer 20, 9 (1987), 17–41. doi:10.1109/MC.1987.1663693

Robbe Cools, Jihae Han, and Adalberto L. Simeone. 2021. SelectVisAR: Selective Visualisation of Virtual Environments in Augmented Reality. In Proceedings of the 2021 ACM Designing Interactive Systems Conference (Virtual Event, USA) (DIS '21). Association for Computing Machinery, New York, NY, USA, 275–282. doi:10.1145/3461778.3462096

DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, Bing Xue, Bingxuan Wang, Bochao Wu, Bei Feng, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, Damai Dai, Deli Chen, Dongjie Ji, Erhang Li, Fangyun Lin, Fucong Dai, Fuli Luo, Guangbo Hao, Guanting Chen, Guowei Li, H. Zhang, Han Bao, Hanwei Xu, Haocheng Wang, Honghui Ding, Huajian Xin, Huazuo Gao, Hui Qu, Hui Li, Jianzhong Guo, Jiashi Li, Jiawei Wang, Jingchang Chen, Jingyang Yuan, Junjie Qiu, Junlong Li, J. L. Cai, Jiaqi Ni, Jian Liang, Jin Chen, Kai Dong, Kai Hu, Kaige Gao, Kang Guan, Kexin Huang, Kuai Yu, Lean Wang, Lecong Zhang, Liang Zhao, Litong Wang, Liyue Zhang, Lei Xu, Leyi Xia, Mingchuan Zhang, Minghua Zhang, Minghui Tang, Meng Li, Miaojun Wang, Mingming Li, Ning Tian, Panpan Huang, Peng Zhang, Qiancheng Wang, Qinyu Chen, Qiushi Du, Ruiqi Ge, Ruisong Zhang, Ruizhe Pan, Runji Wang, R. J. Chen, R. L. Jin, Ruyi Chen, Shanghao Lu, Shangyan Zhou, Shanhuang Chen, Shengfeng Ye, Shiyu Wang, Shuiping Yu, Shunfeng Zhou, Shuting Pan, S. S. Li, Shuang Zhou, Shaoqing Wu, Shengfeng Ye, Tao Yun, Tian Pei, Tianyu Sun, T. Wang, Wangding Zeng, Wanjia Zhao, Wen Liu, Wenfeng Liang, Wenjun Gao, Wenqin Yu, Wentao Zhang, W. L. Xiao, Wei An, Xiaodong Liu, Xiaohan Wang, Xiaokang Chen, Xiaotao Nie, Xin Cheng, Xin Liu, Xin Xie, Xingchao Liu, Xinyu Yang, Xinyuan Li, Xuecheng Su, Xuheng Lin, X. Q. Li, Xiangyue Jin, Xiaojin Shen, Xiaosha Chen, Xiaowen Sun, Xiaoxiang Wang, Xinnan Song, Xinyi Zhou, Xianzu Wang, Xinxia Shan, Y. K. Li, Y. Q. Wang, Y. X. Wei, Yang Zhang, Yanhong Xu, Yao Li, Yao Zhao, Yaofeng Sun, Yaohui Wang, Yi Yu, Yichao Zhang, Yifan Shi, Yiliang Xiong, Ying He, Yishi Piao, Yisong Wang, Yixuan Tan, Yiyang Ma, Yiyuan Liu, Yongqiang Guo, Yuan Ou, Yuduan Wang, Yue Gong, Yuheng Zou, Yujia He, Yunfan Xiong, Yuxiang Luo, Yuxiang You, Yuxuan Liu, Yuyang Zhou, Y. X. Zhu, Yanhong Xu, Yanping Huang, Yaohui Li, Yi Zheng, Yuchen Zhu, Yunxian Ma, Ying Tang, Yukun Zha, Yuting Yan, Z. Z. Ren, Zehui Ren, Zhangli Sha, Zhe Fu, Zhean Xu, Zhenda Xie, Zhengyan Zhang, Zhewen Hao, Zhicheng Ma, Zhigang Yan, Zhiyu Wu, Zihui Gu, Zijia Zhu, Zijun Liu, Zilin Li, Ziwei Xie, Ziyang Song, Zizheng Pan, Zhen Huang, Zhipeng Xu, Zhongyu Zhang, and Zhen Zhang. 2025. DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning. arXiv:2501.12948 [cs.CL] https://arxiv.org/abs/2501.12948

Xiaofei Dong, Xueqiang Zhang, Weixin Bu, Dan Zhang, and Feng Cao. 2024. A Survey of LLM-based Agents: Theories, Technologies, Applications and Suggestions. In 2024 3rd International Conference on Artificial Intelligence, Internet of Things and Cloud Computing Technology (AIoTC). 407–413. doi:10.1109/AIoTC63215.2024.10748304

Yogesh K. Dwivedi, Laurie Hughes, Abdullah M. Baabdullah, Samuel Ribeiro-Navarrete, Mihalis Giannakis, Mutaz M. Al-Debei, Denis Dennehy, Bhimaraya Metri, Dimitrios Buhalis, Christy M. K. Cheung, Kieran Conboy, Ronan Doyle, Rameshwar Dubey, Vincent Dutot, Reto Felix, D. P. Goyal, Anders Gustafsson, Chris Hinsch, Ikram Jebabli, Marijn Janssen, Young-Gab Kim, Jooyoung Kim, Stefan Koos, David Kreps, Nir Kshetri, Vikram Kumar, Keng-Boon Ooi, Savvas Papagiannidis, Ilias O. Pappas, Ariana Polyviou, Sang-Min Park, Neeraj Pandey, Maciel M. Queiroz, Ramakrishnan Raman, Philipp A. Rauschnabel, Anuragini Shirish, Marianna Sigala, Konstantina Spanaki, Garry Wei-Han Tan, Manoj Kumar Tiwari, Giampaolo Viglia, and Samuel Fosso Wamba. 2022. Metaverse beyond the hype: Multidisciplinary perspectives on emerging challenges, opportunities, and agenda for research, practice and policy. Int. J. Inf. Manag. 66, C (10 2022), 55 pages. doi:10.1016/j.ijinfomgt.2022.102542

Tomaž Erjavec, Matyáš Kopp, Nikola Ljubešić, Taja Kuzman, Paul Rayson, Petya Osenova, Maciej Ogrodniczuk, Çağrı Çöltekin, Danijel Koržinek, Katja Meden, Jure Skubic, Peter Rupnik, Tommaso Agnoloni, José Aires, Starkaður Barkarson, Roberto Bartolini, Núria Bel, María Calzada Pérez, Roberts Darģis, Sascha Diwersy, Maria Gavriilidou, Ruben van Heusden, Mikel Iruskieta, Neeme Kahusk, Anna Kryvenko, Noémi Ligeti-Nagy, Carmen Magariños, Martin Mölder, Costanza Navarretta, Kiril Simov, Lars Magne Tungland, Jouni Tuominen, John Vidler, Adina Ioana Vladu, Tanja Wissik, Väinö Yrjänäinen, and Darja Fišer. 2024. ParlaMint II: advancing comparable parliamentary corpora across Europe. Language Resources and Evaluation (Dec. 2024), 32 pages. doi:10.1007/s10579-024-09798-w

Tomaž Erjavec, Maciej Ogrodniczuk, Petya Osenova, Nikola Ljubešić, Kiril Simov, Andrej Pančur, Michał Rudolf, Matyáš Kopp, Starkaður Barkarson, Steinþór Steingrímsson, Çağrı Çöltekin, Jesse de Does, Katrien Depuydt, Tommaso Agnoloni, Giulia Venturi, María Calzada Pérez, Luciana D. de Macedo, Costanza Navarretta, Giancarlo Luxardo, Matthew Coole, Paul Rayson, Vaidas Morkevičius, Tomas Krilavičius, Roberts Darģis, Orsolya Ring, Ruben van Heusden, Maarten Marx, and Darja Fišer. 2022. The ParlaMint Corpora of Parliamentary Proceedings. Lang. Resour. Eval. 57, 1 (feb 2022), 415–448. doi:10.1007/s10579-021-09574-0

David Ferrucci, Adam Lally, Karin Verspoor, and Eric Nyberg. 2009. Unstructured Information Management Architecture (UIMA) Version 1.0. OASIS Standard. https://docs.oasis-open.org/uima/v1.0/uima-v1.0.html

Sebastian J Friston, Ben J Congdon, David Swapp, Lisa Izzouzi, Klara Brandstätter, Daniel Archer, Otto Olkkonen, Felix Johannes Thiel, and Anthony Steed. 2021. Ubiq: A System to Build Flexible Social Virtual Reality Experiences. In Proceedings of the 27th ACM Symposium on Virtual Reality Software and Technology (Osaka, Japan) (VRST '21), Yuichi Itoh, Kazuki Takashima, Parinya Punpongsanon, Misha Sra, Kazuyuki Fujita, Shigeo Yoshida, and Piumsomboon. Tham (Eds.). Association for Computing Machinery, New York, NY, USA, Article 6, 11 pages. doi:10.1145/3489849.3489871

S. Furui, T. Kikuchi, Y. Shinnaka, and C. Hori. 2004. Speech-to-text and speech-to-speech summarization of spontaneous speech. IEEE Transactions on Speech and Audio Processing 12, 4 (2004), 401–408. doi:10.1109/TSA.2004.828699

Giuliano Gaia, Stefania Boiano, and Ann Borda. 2019. Engaging Museum Visitors with AI: The Case of Chatbots. Springer International Publishing, Cham, 309–329. doi:10.1007/978-3-319-97457-615

Jens Grubert. 2023. Mixed Reality Interaction Techniques. Springer International Publishing, Cham, 109–129. doi:10.1007/978-3-030-67822-75

Ryan Richard H. Guadana, Justin Macki B. Julhusin, Czarisse Gem F. Bucas, Kaycee Kate B. Ferrer, and Dan Rafael M. Domingo. 2024. iVenue: A Framework for 3D Virtual Tour Experience with Chatbot for Philippine Universities. In Proceedings of the 2024 15th International Conference on E-Education, E-Business, E-Management and E-Learning (Fukuoka-shi, Japan) (IC4E '24). Association for Computing Machinery, New York, NY, USA, 442–449. doi:10.1145/3670013.3670073

Zhicheng Guo, Sijie Cheng, Yile Wang, Peng Li, and Yang Liu. 2023. Prompt-Guided Retrieval Augmentation for Non-Knowledge-Intensive Tasks. arXiv:2305.17653 [cs.CL] https://arxiv.org/abs/2305.17653

Charlie Hargood, Verity Hunt, Mark J. Weal, and David E. Millard. 2016. Patterns of Sculptural Hypertext in Location Based Narratives. In Proceedings of the 27th ACM Conference on Hypertext and Social Media (Halifax, Nova Scotia, Canada) (HT '16). Association for Computing Machinery, New York, NY, USA, 61–70. doi:10.1145/2914586.2914595

Ziyao He, Shiyuan Li, Yunpeng Song, and Zhongmin Cai. 2024. Towards Building Condition-Based Cross-Modality Intention-Aware Human-AI Cooperation under VR Environment. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI '24). Association for Computing Machinery, New York, NY, USA, Article 955, 13 pages. doi:10.1145/3613904.3642360

Matthew Honnibal, Ines Montani, Sofie Van Landeghem, and Adriane Boyd. 2020. spaCy: Industrial-strength Natural Language Processing in Python. doi:10.5281/zenodo.1212303

Juan Izquierdo-Domenech, Jordi Linares-Pellicer, and Isabel Ferri-Molla. 2024. Virtual Reality and Language Models, a New Frontier in Learning. International Journal of Interactive Multimedia and Artificial Intelligence 8, 5 (2024), 46 pages. doi:10.9781/ijimai.2024.02.007

David Kolb. 2013. Adaptive Hypertext Narrative As City Planning. In Proceedings of the 24th ACM Conference on Hypertext and Social Media (Paris, France) (HT '13). ACM, New York, NY, USA, 184–188. doi:10.1145/2481492.2481514

Elif Hilal Korkut and Elif Surer. 2023. Visualization in virtual reality: a systematic review. Virtual Reality 27, 2 (Jan. 2023), 1447–1480. doi:10.1007/s10055-023-00753-8

George p. Landow. 1987. Relationally Encoded Links and the Rhetoric of Hypertext. In Proceedings of the ACM Conference on Hypertext (Chapel Hill, North Carolina, USA) (HYPERTEXT '87). Association for Computing Machinery, New York, NY, USA, 331–343. doi:10.1145/317426.317450

Alexander Leonhardt, Giuseppe Abrami, Daniel Baumartz, and Alexander Mehler. 2023. Unlocking the Heterogeneous Landscape of Big Data NLP with DUUI. In Findings of the Association for Computational Linguistics: EMNLP 2023, Houda Bouamor, Juan Pino, and Kalika Bali (Eds.). Association for Computational Linguistics, Singapore, 385–399. https://aclanthology.org/2023.findings-emnlp.29

Ziming Li, Pinaki Prasanna Babar, Mike Barry, and Roshan L Peiris. 2024. Exploring the Use of Large Language Model-Driven Chatbots in Virtual Reality to Train Autistic Individuals in Job Communication Skills. In Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI EA '24). Association for Computing Machinery, New York, NY, USA, Article 156, 7 pages. doi:10.1145/3613905.3651996

Aislyn P. C. Lin, Charles V. Trappey, Chi-Cheng Luan, Amy J. C. Trappey, and Kevin L. K. Tu. 2021. A Test Platform for Managing School Stress Using a Virtual Reality Group Chatbot Counseling System. Applied Sciences 11, 19 (2021), 24 pages. doi:10.3390/app11199071

Jonathan Linowes. 2015. Unity virtual reality projects. Packt Publishing Ltd, Birmingham.

Stefan Marks, Javier E. Estevez, and Andy M. Connor. 2014. Towards the Holodeck: Fully Immersive Virtual Reality Visualisation of Scientific and Engineering Data. In Proceedings of the 29th International Conference on Image and Vision Computing New Zealand (Hamilton, New Zealand) (IVCNZ '14). Association for Computing Machinery, New York, NY, USA, 42–47. doi:10.1145/2683405.2683424

Kieran W. May, James Walsh, Ross T. Smith, Ning Gu, and Bruce H. Thomas. 2023. Examining VR Technologies for Immersive Lighting Performance Simulation and Visualisation in Building Design and Analysis. In 2023 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW). IEEE, 913–914. doi:10.1109/VRW58643.2023.00299

Hugo Mayo, Alastair Shipman, Daniele Giunchi, Riccardo Bovo, Anthony Steed, and Thomas Heinis. 2022. VR Toolkit for Identifying Group Characteristics. Collective Dynamics 6 (Feb. 2022), 1–17. doi:10.17815/cd.2021.119

Alexander Mehler, Mevlüt Bagci, Alexander Henlein, Giuseppe Abrami, Christian Spiekermann, Patrick Schrottenbacher, Maxim Konca, Andy Lücking, Juliane Engel, Marc Quintino, Jakob Schreiber, Kevin Saukel, and Olga Zlatkin-Troitschanskaia. 2023. A Multimodal Data Model for Simulation-Based Learning with Va.Si.Li-Lab. In Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management, Vincent G. Duffy (Ed.). Springer Nature Switzerland, Cham, 539–565. doi:10.1007/978-3-031-35741-139

Thorsten Müller and Dominik Kreutz. 2021. Thorsten-Voice Dataset 2021.02. doi:10.5281/zenodo.5525342 Please use it to make the world a better place for whole humankind..

Peter J. Nürnberg, Erich R. Schneider, and John J. Leggett. 1996. Designing Digital Libraries for the Hyperliterate Age. JUCS - Journal of Universal Computer Science 2, 9 (1996), 610–622. arXiv:https://doi.org/10.3217/jucs-002-09-0610 doi:10.3217/jucs-002-09-0610

OpenAI. 2024. TTS OpenAI Text to Speech. https://platform.openai.com/docs/guides/text-to-speech/overview

Mengxu Pan, Alexandra Kitson, Hongyu Wan, and Mirjana Prpa. 2024. ELLMA-T: an Embodied LLM-agent for Supporting English Language Learning in Social VR. arXiv:2410.02406 [cs.HC] https://arxiv.org/abs/2410.02406

Mariusz Pisarski. 2024. Hyperlepis - a secret weapon of hypertext rhetoric. In Proceedings of the 35th ACM Conference on Hypertext and Social Media (Poznan, Poland) (HT '24). Association for Computing Machinery, New York, NY, USA, 203–207. doi:10.1145/3648188.3675149

Alex Prior and Maanasa Sivashankar. 2023. Our future in space: the physical and virtual opening-up of parliaments to publics. The Journal of Legislative Studies 29, 3 (2023), 463–481. arXiv:https://doi.org/10.1080/13572334.2023.2202090 doi:10.1080/13572334.2023.2202090

C. Pybus, K. Graham, J. Doherty, N. Arellano, and S. Fai. 2019. NEW REALITIES FOR CANADA'S PARLIAMENT: A WORKFLOW FOR PREPARING HERITAGE BIM FOR GAME ENGINES AND VIRTUAL REALITY. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-2/W15 (2019), 945–952. doi:10.5194/isprs-archives-XLII-2-W15-945-2019

Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, and Ilya Sutskever. 2023. Robust speech recognition via large-scale weak supervision. In Proceedings of the 40th International Conference on Machine Learning (Honolulu, Hawaii, USA) (ICML'23). JMLR.org, Article 1182, 27 pages. doi:10.5555/3618408.3619590

M. Z. Rashad, Hazem M. El-Bakry, Islam R. Isma'il, and Nikos Mastorakis. 2010. An overview of text-to-speech synthesis techniques. In Proceedings of the 4th International Conference on Communications and Information Technology (Corfu Island, Greece) (CIT'10). World Scientific and Engineering Academy and Society (WSEAS), Stevens Point, Wisconsin, USA, 84–89.

Philipp A. Rauschnabel, Reto Felix, Chris Hinsch, Hamza Shahab, and Florian Alt. 2022. What is XR? Towards a Framework for Augmented and Virtual Reality. Computers in Human Behavior 133 (2022), 107289. doi:10.1016/j.chb.2022.107289

Asreen Rostami, Chiara Rossitto, and Annika Waern. 2018. Frictional Realities: Enabling Immersion in Mixed-Reality Performances. In Proceedings of the 2018 ACM International Conference on Interactive Experiences for TV and Online Video (SEOUL, Republic of Korea) (TVX '18). Association for Computing Machinery, New York, NY, USA, 15–27. doi:10.1145/3210825.3210827

Simon Rowberry. 2023. Historiographies of Hypertext. In Proceedings of the 34th ACM Conference on Hypertext and Social Media (Rome, Italy) (HT '23). Association for Computing Machinery, New York, NY, USA, Article 41, 10 pages. doi:10.1145/3603163.3609038

Thomas Schedel and Claus Atzenbeck. 2016. Spatio-Temporal Parsing in Spatial Hypermedia. In Proceedings of the 27th ACM Conference on Hypertext and Social Media (Halifax, Nova Scotia, Canada) (HT '16). Association for Computing Machinery, New York, NY, USA, 149–157. doi:10.1145/2914586.2914596

Donghoon Shin, Soomin Kim, Ruoxi Shang, Joonhwan Lee, and Gary Hsieh. 2023. IntroBot: Exploring the Use of Chatbot-assisted Familiarization in Online Collaborative Groups. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems (Hamburg, Germany) (CHI '23). Association for Computing Machinery, New York, NY, USA, Article 613, 13 pages. doi:10.1145/3544548.3580930

Frank M. Shipman and Catherine C. Marshall. 1999. Spatial Hypertext: An Alternative to Navigational and Semantic Links. Comput. Surveys 31, 4 (dec 1999), 14–es. doi:10.1145/345966.346001

João Simões, Anderson Maciel, Catarina Moreira, Maurício Sousa, and Joaquim Jorge. 2025. SPARC: Shared Perspective with Avatar Distortion for Remote Collaboration in VR. In Advances in Computer Graphics, Nadia Magnenat-Thalmann, Jinman Kim, Bin Sheng, Zhigang Deng, Daniel Thalmann, and Ping Li (Eds.). Springer Nature Switzerland, Cham, 99–112.

Pauline A. Smith and John R. Wilson. 1993. Navigation in hypertext through virtual environments. Applied Ergonomics 24, 4 (1993), 271–278. doi:10.1016/0003-6870(93)90462-I

Paul Smolensky, Brigham Bell, Barbara Fox, Roger King, and Clayton Lewis. 1987. Constraint-Based Hypertext for Argumentation. In Proceedings of the ACM Conference on Hypertext (Chapel Hill, North Carolina, USA) (HYPERTEXT '87). Association for Computing Machinery, New York, NY, USA, 215–245. doi:10.1145/317426.317443

Saba Sturua, Isabelle Mohr, Mohammad Kalim Akram, Michael Günther, Bo Wang, Markus Krimmel, Feng Wang, Georgios Mastrapas, Andreas Koukounas, Nan Wang, and Han Xiao. 2024. jina-embeddings-v3: Multilingual Embeddings With Task LoRA. arXiv:2409.10173 [cs.CL] https://arxiv.org/abs/2409.10173

Karsten Tymann, Matthias Lutz, Patrick Palsbröker, and Carsten Gips. 2019. GerVADER - A German Adaptation of the VADER Sentiment Analysis Tool for Social Media Texts. In Proceedings of the Conference on "Lernen, Wissen, Daten, Analysen", Berlin, Germany, September 30 - October 2, 2019 (CEUR Workshop Proceedings, Vol. 2454), Robert Jäschke and Matthias Weidlich (Eds.). CEUR-WS.org, 178–189. https://ceur-ws.org/Vol-2454/paper14.pdf

Stephen Uzor and Per Ola Kristensson. 2021. An Exploration of Freehand Crossing Selection in Head-Mounted Augmented Reality. ACM Trans. Comput.-Hum. Interact. 28, 5, Article 33 (Aug. 2021), 27 pages. doi:10.1145/3462546

Savvas Varitimiadis, Konstantinos Kotis, Andreas Skamagis, Alexandros Tzortzakakis, George Tsekouras, and Dimitris Spiliotopoulos. 2020. Towards implementing an AI chatbot platform for museums. International Conference on Cultural Informatics, Communication & Media Studies 1, 1 (2020), 15 pages. doi:10.12681/cicms.2732

Hongyu Wan, Jinda Zhang, Abdulaziz Arif Suria, Bingsheng Yao, Dakuo Wang, Yvonne Coady, and Mirjana Prpa. 2024. Building LLM-based AI Agents in Social Virtual Reality. In Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (Honolulu, HI, USA) (CHI EA '24). Association for Computing Machinery, New York, NY, USA, Article 65, 7 pages. doi:10.1145/3613905.3651026

Yiwen Xu, Qinyang Hou, Hongyu Wan, and Mirjana Prpa. 2024. Safe Guard: an LLM-agent for Real-time Voice-based Hate Speech Detection in Social Virtual Reality. arXiv:2409.15623 [eess.AS] https://arxiv.org/abs/2409.15623