Strategies in Transfer Learning for Low-Resource Speech Synthesis: Phone Mapping, Features Input, and Source Language Selection

Spremljeno u:

Bibliografski detalji
Izdano u:	arXiv.org (Jun 21, 2023), p. n/a
Glavni autor:	Do, Phat
Daljnji autori:	Coler, Matt, Dijkstra, Jelske, Klabbers, Esther
Izdano:	Cornell University Library, arXiv.org
Teme:	Mapping Learning Automatic speech recognition Languages Family trees
Online pristup:	Citation/Abstract Full text outside of ProQuest
Oznake:	Dodaj oznaku Bez oznaka, Budi prvi tko označuje ovaj zapis!

Opis
Sažetak:	We compare using a PHOIBLE-based phone mapping method and using phonological features input in transfer learning for TTS in low-resource languages. We use diverse source languages (English, Finnish, Hindi, Japanese, and Russian) and target languages (Bulgarian, Georgian, Kazakh, Swahili, Urdu, and Uzbek) to test the language-independence of the methods and enhance the findings' applicability. We use Character Error Rates from automatic speech recognition and predicted Mean Opinion Scores for evaluation. Results show that both phone mapping and features input improve the output quality and the latter performs better, but these effects also depend on the specific language combination. We also compare the recently-proposed Angular Similarity of Phone Frequencies (ASPF) with a family tree-based distance measure as a criterion to select source languages in transfer learning. ASPF proves effective if label-based phone input is used, while the language distance does not have expected effects.
ISSN:	2331-8422
Izvor:	Engineering Database