Enhancing E-Book Reading: Integration Text to Speech and BART Model Summarization
Reading is the main foundation of the continuous learning process and a universal right for every individual. Reading activities provide access to new knowledge, broaden the mind, and broaden horizons. Along with technological advances, digital books or e-books have become an important alternative to traditional printed books. E-books not only provide text but also images, videos, and audio that can be accessed through electronic devices. To maximize the potential of e-books, integration of technologies such as text-to-speech systems and text summarization is needed. The text-to-speech system allows users to listen to e-book content, while text summarization using the BART method conveys complex information without losing its essence. The application of this technology on the AllInclusiveReads platform is expected to help people who have limitations in visual abilities or reading long texts and can improve the user's reading experience. Considering the test findings with the evaluation method, the best result was obtained in the calculation of the Rouge 1 monogram overlap, which was 0.5735 for the Precision value, 0.4615 for the recall value and 0.5477 for the F1 score value. This research will be developed in the future so that it can not only process PDF files, but also other formats. This research will be developed in the future using other methods, so that it can not only process PDF document files, but also other formats.
Authors
- Agung Gumelar (ORCID: https://orcid.org/0009-0006-2855-4189)
- Ivana Lucia Kharisma (ORCID: https://orcid.org/0009-0000-8017-1579)
- Indra Yustiana
- Kamdan Kamdan (ORCID: https://orcid.org/0009-0000-8468-7716)
- Satria Rizki Rohmat
- Moh. Saeful Rohman
Institutions
- Universitas Wijaya Putra (ID)
- Nusa Putra University (ID)
Publication Details
- Published
- 2024-12-11
- DOI
- https://doi.org/10.1109/icced64257.2024.10983069
- Citations
- 1
- Primary Topic
- Mathematics, Computing, and Information Processing
- Type
- article
- Field-Weighted Citation Impact
- 0.36