Az MI által generált beszéd jellemzői
AI
DOI :
https://doi.org/10.46522/S.2024.02.6Mots-clés :
speech, AI, speech synthesis applicationsRésumé
Characteristics of AI-Generated Speech
Success in life is not primarily attributed to cognitive skills. Rather, it is the level of development in social skills that creates real connections, forms relationships, and keeps attention engaged. There are a few human abilities that, for the time being, can still compete with artificial intelligence: our capacity for critical thinking, and our intelligence for dealing with human emotions and relationships. Machines are better at active attention, as they are able to pay attention continuously, without interruption. In our attention-deficient society, this is a fact of prime importance. Speech generated by AI can vocalize written text and segment it according to punctuation marks, but it has not yet mastered the vocal exploration of deeper interpretive connections. Will it be capable of doing so in the future? Let’s examine the vocal characteristics of human speech and draw parallels with the opportunities offered by AI-generated speech synthesis applications! With the advancement of human robots, personal and technological developments will make speech modulations, the creation of individual tones, and the feeding of memories a part of the algorithm in the near future.
Références
KURZWEIL, R., 2022. Hogyan alkossunk elmét? Budapest: Pallas Athéné Books.
NIKLÉCZY, P. és OLASZY, G., 2016. Kempelen Farkas beszélőgépének rekonstrukciója. Nyelvtudományi Kutatóközpont, No. 4.
TILESCH, Gy. és HATAMLEH, O., 2021. Mesterséges intelligencia. Budapest: Libri Kiadó.
Téléchargements
Publiée
Comment citer
Numéro
Rubrique
Licence
Ce travail est disponible sous la licence Creative Commons Attribution 4.0 International .
This work is licensed under a CC BY Creative Commons Attribution 4.0 International License, which permits any use, reproduction, distribution, self-archiving and citation of the work as long as the authors are credited. The complete bibliographical data of Symbolon Journal must also be indicated, which you can find in the How to cite section on this page. If possible, please also place a link leading to the original publication. Copyright of articles belongs to the authors.