The Use of Artificial Intelligence During Presentations: A Review on Quality and Fluency
Yıl 2025,
Cilt: 10 Sayı: 10, 126 - 144, 25.03.2025
Mehmet Göktuğ Gökçe
,
Asli Şemşimoğlu
,
Sinan Malak
,
Rabia Durgut
Öz
Nowadays, presentations are widely used as a crucial communication tool in both business and educational settings. At this stage, the quality and fluency of presentations play a significant role in the audience’s comprehension and engagement. However, speakers may sometimes encounter unexpected challenges during presentations. Such challenges may include forgetting or skipping topics, possibly due to anxiety or excitement. This study aims to enhance the quality and fluency of presentations using natural language processing (NLP) techniques, a subfield of artificial intelligence. In addition to fine-tuning large language models (LLMs), which are one of the application areas of NLP, text-to-speech (TTS) and speech-to-text (STT) conversion techniques will also be employed. The dataset used in this study was synthetically generated, and test results have been obtained through different large language models. Efforts continue to improve accuracy through fine-tuning methods. The results indicate that leveraging large language models can enhance presentation performance.
Kaynakça
- Agrawal, D., & Itankar, P. (2020). Real-time contextual searches to assist speakers. Technical Disclosure Commons, 29 Ekim 2020.
- Arriaga, E. M. (2017). Automatic slide progression during a presentation. Technical Disclosure Commons, 5 Aralık 2017.
- Bachmann, M., Subramaniam, A., Born, J., & Weibel, D. (2023). Virtual reality public speaking training: effectiveness and user technology acceptance. Frontiers in Virtual Reality, 4. https://doi.org/10.3389/frvir.2023.1242544
- Fujimoto, Y., Hangyu, Z., Sawabe, T., Kanbara, M., & Kato, H. (2023). Stop bad real-time feedback!: Estimation of the timing of feedback that negatively impacts presenters for presentation training in virtual reality. 2023 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct), 405–410. https://doi.org/10.1109/ismar-adjunct60411.2023.00087
- Hraška, P., Drobny, L., Cechovsky, J., Gregor, G., & Šuppa, M. (2023). Real-time audience analytics system for measuring engagement and sentiment during live presentations. Technical Disclosure Commons, 27 Haziran 2023.
- Inoue, R., Shiramatsu, S., Ozono, T., & Shintani, T. (2014). Visualizing real-time questionnaire results to promote participation in interactive presentations. 2014 IIAI 3rd International Conference on Advanced Applied Informatics, 64–69. https://doi.org/10.1109/iiai-aai.2014.24
- Kimani, E., Bickmore, T., Trinh, H., & Pedrelli, P. (2019). You’ll be great: Virtual agent-based cognitive restructuring to reduce public speaking anxiety. 2019 8th International Conference on Affective Computing and Intelligent Interaction (ACII). https://doi.org/10.1109/acii.2019.8925438
- Kurihara, K., Goto, M., Ogata, J., Matsusaka, Y., & Igarashi, T. (2007). Presentation sensei. Proceedings of the 9th International Conference on Multimodal Interfaces, 358–365. https://doi.org/10.1145/1322192.1322256
- Mentimeter. (2024). Mentimeter platformu. https://www.mentimeter.com/ (Son güncelleme: 29 Nisan 2024).
- Powtoon. (2024). Powtoon platformu. https://www.powtoon.com/ (Erişim tarihi: 13 Kasım 2024).
- Teja, D., & Sedouram, R. (2023). Improving online presentations based on audience usage of conversational assistant. Technical Disclosure Commons, 8 Aralık 2023.
- Trinh, H., Asadi, R., Edge, D., & Bickmore, T. (2017). RoboCOP. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 1(2), 1–24. https://doi.org/10.1145/3090092
- Yi, S., Yumoto, H., Wang, X., & Yamasaki, T. (2020). Presentationtrainer: Oral presentation support system for impression-related feedback. Proceedings of the AAAI Conference on Artificial Intelligence, 34(09), 13644–13645. https://doi.org/10.1609/aaai.v34i09.7109
- Zhao, X., & Fei, F. (2022). Investigation on the design of anthropomorphic oral presentation assistant training system. Mobile Information Systems, 2022, 4-10. https://doi.org/10.1155/2022/3719010
Sunum Esnasında Yapay Zeka Kullanımı: Kalite ve Akıcılık Üzerine Bir İnceleme
Yıl 2025,
Cilt: 10 Sayı: 10, 126 - 144, 25.03.2025
Mehmet Göktuğ Gökçe
,
Asli Şemşimoğlu
,
Sinan Malak
,
Rabia Durgut
Öz
Günümüzde sunumlar, iş dünyasında ve eğitim ortamlarında önemli bir iletişim aracı olarak kullanılmaktadır. Bu aşamada sunumların kalitesi ve akıcılığı dinleyicilerin sunumdan aldığı verimi için büyük rol oynamaktadır. Ancak konuşmacılar bazen sunum esnasında beklenmedik olumsuzluklarla karşılaşabilmektedirler. Söz konusu olumsuzluklar arasında, muhtemel olarak anksiyete veya heyecan sebebiyle önemli noktaların unutulması ya da atlanması gibi durumlar örnek gösterilebilir. Çalışmada yapay zekanın bir alt alanı olan doğal dil işleme tekniklerinin kullanımı ile sunum kalitesinin ve akıcılığının iyileştirilebilmesi üzerine çalışılacaktır. Doğal dil işlemenin uygulama alanlarından biri olan büyük dil modelleri (LLMs) üzerinde ince ayar (fine-tune) yapılmasının yanı sıra, metinden sese (Text-to-Speech) ve sesten metne (Speech-to-Text) çevirme teknikleri de kullanılacaktır. Çalışmada yer alan veri seti sentetik olarak hazırlanmış olmakla birlikte farklı büyük dil modelleri üzerinden test sonuçları elde edilmiştir. Test sonuçlarının, ince ayar yöntemi ile doğruluğunun arttırılması üzerine çalışmalar devam etmektedir. Elde edilen sonuçlar büyük dil modelleri kullanımı sayesinde sunum performansının artırılabileceğini göstermektedir.
Kaynakça
- Agrawal, D., & Itankar, P. (2020). Real-time contextual searches to assist speakers. Technical Disclosure Commons, 29 Ekim 2020.
- Arriaga, E. M. (2017). Automatic slide progression during a presentation. Technical Disclosure Commons, 5 Aralık 2017.
- Bachmann, M., Subramaniam, A., Born, J., & Weibel, D. (2023). Virtual reality public speaking training: effectiveness and user technology acceptance. Frontiers in Virtual Reality, 4. https://doi.org/10.3389/frvir.2023.1242544
- Fujimoto, Y., Hangyu, Z., Sawabe, T., Kanbara, M., & Kato, H. (2023). Stop bad real-time feedback!: Estimation of the timing of feedback that negatively impacts presenters for presentation training in virtual reality. 2023 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct), 405–410. https://doi.org/10.1109/ismar-adjunct60411.2023.00087
- Hraška, P., Drobny, L., Cechovsky, J., Gregor, G., & Šuppa, M. (2023). Real-time audience analytics system for measuring engagement and sentiment during live presentations. Technical Disclosure Commons, 27 Haziran 2023.
- Inoue, R., Shiramatsu, S., Ozono, T., & Shintani, T. (2014). Visualizing real-time questionnaire results to promote participation in interactive presentations. 2014 IIAI 3rd International Conference on Advanced Applied Informatics, 64–69. https://doi.org/10.1109/iiai-aai.2014.24
- Kimani, E., Bickmore, T., Trinh, H., & Pedrelli, P. (2019). You’ll be great: Virtual agent-based cognitive restructuring to reduce public speaking anxiety. 2019 8th International Conference on Affective Computing and Intelligent Interaction (ACII). https://doi.org/10.1109/acii.2019.8925438
- Kurihara, K., Goto, M., Ogata, J., Matsusaka, Y., & Igarashi, T. (2007). Presentation sensei. Proceedings of the 9th International Conference on Multimodal Interfaces, 358–365. https://doi.org/10.1145/1322192.1322256
- Mentimeter. (2024). Mentimeter platformu. https://www.mentimeter.com/ (Son güncelleme: 29 Nisan 2024).
- Powtoon. (2024). Powtoon platformu. https://www.powtoon.com/ (Erişim tarihi: 13 Kasım 2024).
- Teja, D., & Sedouram, R. (2023). Improving online presentations based on audience usage of conversational assistant. Technical Disclosure Commons, 8 Aralık 2023.
- Trinh, H., Asadi, R., Edge, D., & Bickmore, T. (2017). RoboCOP. Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, 1(2), 1–24. https://doi.org/10.1145/3090092
- Yi, S., Yumoto, H., Wang, X., & Yamasaki, T. (2020). Presentationtrainer: Oral presentation support system for impression-related feedback. Proceedings of the AAAI Conference on Artificial Intelligence, 34(09), 13644–13645. https://doi.org/10.1609/aaai.v34i09.7109
- Zhao, X., & Fei, F. (2022). Investigation on the design of anthropomorphic oral presentation assistant training system. Mobile Information Systems, 2022, 4-10. https://doi.org/10.1155/2022/3719010