The Effects of Input Type and Pronunciation Dictionary Usage in Transfer Learning for Low-Resource Text-to-Speech

Guardado en:
Detalles Bibliográficos
Publicado en:arXiv.org (Jun 1, 2023), p. n/a
Autor principal: Do, Phat
Otros Autores: Coler, Matt, Dijkstra, Jelske, Klabbers, Esther
Publicado:
Cornell University Library, arXiv.org
Materias:
Acceso en línea:Citation/Abstract
Full text outside of ProQuest
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
Descripción
Resumen:We compare phone labels and articulatory features as input for cross-lingual transfer learning in text-to-speech (TTS) for low-resource languages (LRLs). Experiments with FastSpeech 2 and the LRL West Frisian show that using articulatory features outperformed using phone labels in both intelligibility and naturalness. For LRLs without pronunciation dictionaries, we propose two novel approaches: a) using a massively multilingual model to convert grapheme-to-phone (G2P) in both training and synthesizing, and b) using a universal phone recognizer to create a makeshift dictionary. Results show that the G2P approach performs largely on par with using a ground-truth dictionary and the phone recognition approach, while performing generally worse, remains a viable option for LRLs less suitable for the G2P approach. Within each approach, using articulatory features as input outperforms using phone labels.
ISSN:2331-8422
Fuente:Engineering Database