EUVIMEDEUVIMED entdecken
Crossref

Which One Sounds More Human-like? —Comparison of Synthetic Speech Trained on Japanese Daily Conversational Data with and without Disfluency

Akiko Mokhtari, Hiroaki Hatano, Jun Arai, Nick Campbell, Toshiyuki Sadanobu

Journal of Natural Language Processing · 2026 · 33 (1) · 186-206

DOI: 10.5715/jnlp.33.186 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Natural Language Processing, 2026. Verfasst von Akiko Mokhtari und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 21 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

As real-world conversational speech is full of disfluencies, it is vital to include them when aiming to create a speech synthesis that reflects real human interactions. While recent studies have developed models capable of automatically inserting disfluencies, this study aims to clarify how five specific types of manually inserted disfluencies commonly observed in Japanese conversational speech—fillers, word-internal prolongations, phrase-final abrupt rising intonation, phrase-final abrupt rising-and-falling intonation, and phrase-final copulas with sentence-final particles—affect the perceived human-likeness of synthetic speech. First, we train an existing AI-based speech synthesis system using daily conversational data from a female Japanese speaker in the Kinki dialect. We then synthesised speech of the same linguistic content with and without disfluencies and asked the survey respondents which sample sounded more human-like. The results showed that a significantly higher percentage of respondents evaluated disfluent synthetic speech as more human-like. We also analysed the relationship between the answers and the respondents' gender, age, and place of upbringing to investigate whether any of these attributes influenced the tendency of the answers. The results suggest that there is an association between the respondents' answers and the regions where they grew up, particularly in response to specific stimulus sounds.

Zitieren

APA

Mokhtari, A., Hatano, H., Arai, J., Campbell, N., & Sadanobu, T. (2026). Which One Sounds More Human-like? —Comparison of Synthetic Speech Trained on Japanese Daily Conversational Data with and without Disfluency. Journal of Natural Language Processing, 33(1), 186-206. https://doi.org/10.5715/jnlp.33.186

Vancouver

Mokhtari A, Hatano H, Arai J, Campbell N, Sadanobu T. Which One Sounds More Human-like? —Comparison of Synthetic Speech Trained on Japanese Daily Conversational Data with and without Disfluency. Journal of Natural Language Processing. 2026;33(1):186-206. doi:10.5715/jnlp.33.186

BibTeX
@article{Mokhtari2026,
  author = {Mokhtari, Akiko and Hatano, Hiroaki and Arai, Jun and Campbell, Nick and Sadanobu, Toshiyuki},
  title = {Which One Sounds More Human-like? —Comparison of Synthetic Speech Trained on Japanese Daily Conversational Data with and without Disfluency},
  journal = {Journal of Natural Language Processing},
  year = {2026},
  volume = {33},
  number = {1},
  pages = {186--206},
  doi = {10.5715/jnlp.33.186}
}
Kostenlos starten →