Usability Of Text-To-Speech Technology in Creating News Podcasts using Portuguese Of Portugal
Authors:
Open Access
- 21 June 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 363-368
Abstract
The increase in the consumption of digital formats has, in many cases, been a penalty for traditional media companies. In the adaptation to digital, the transformation of written news into audio formats, that guarantee spatio-temporal flexibility in its consumption, is one of the differentiating options. Artificial intelligence tools can help accelerate and automate the digitalization processes. It is, therefore, the objective of this paper to evaluate the integration of Text-to-Speech (TTS) technology in the process of creating news podcasts. The study comprised two surveys. The first corresponding to the validation of TTS services in Portuguese from Portugal, and the second for the validation of three models of news podcasts containing human voice, synthesized voice via TTS, and a hybrid model with TTS voice and human voice. The results point to a general acceptance of the integration of voices generated by TTS in news podcasts without prejudice to the consumer experience.Keywords
This publication has 4 references indexed in Scilit:
- Choice of Voices2020
- Speech Synthesis Evaluation — State-of-the-Art Assessment and Suggestion for a Novel Research Program2019
- Text-to-Speech Software and Learning: Investigating the Relevancy of the Voice EffectJournal of Educational Computing Research, 2018
- Beyond the Listening Test: An Interactive Approach to TTS Evaluation2017