Prosody and speech synthesis: an integrative literature review

• Autor(a) principal: Galdino, Julio Cesar
• Data de Publicação: 2023
• Outros Autores: Oliveira Jr., Miguel
• Tipo de documento: Artigo
• Idioma: por
• Título da fonte: Revista da ABRALIN (Online)
• Texto Completo: https://revista.abralin.org/index.php/abralin/article/view/2130
• topic: Síntese de fala; Prosódia; Frequência fundamental; Speech Synthesis; Prosody; Fundamental Frequency

This article aims to present an integrative review of prosody and speech synthesis. To achieve this objective, we elaborated the research question “Which prosodic characteristics are most involved in the improvement of speech synthesis?” and we performed a search on Google Scholar, based on the syntax (prosódia OR entoação OR “frequência fundamental”) AND ("text-to-speech" OR TTS OR "síntese de fala" OR "síntese da fala"). We included 10 studies between 2010 and 2021, which showed that fundamental frequency and pitch are the most expressive features, although text-to-speech systems use other prosodic features to generate synthetic voice intonation or to improve their performance. Furthermore, the results of this review showed that there are still few studies in Brazil on the relationship between prosody and speech synthesis and that joint research between researchers in the fields of linguistics and engineering is important in order to improve the speech synthesis.