Comparative Evaluation of Expressive Japanese Character Text-to-Speech with VITS and Style-BERT-VITS2
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914175403425792 |
|---|---|
| author | Rackauckas, Zackary Hirschberg, Julia |
| author_facet | Rackauckas, Zackary Hirschberg, Julia |
| contents | Synthesizing expressive Japanese character speech poses unique challenges due to pitch-accent sensitivity and stylistic variability. This paper empirically evaluates two open-source text-to-speech models--VITS and Style-BERT-VITS2 JP Extra (SBV2JE)--on in-domain, character-driven Japanese speech. Using three character-specific datasets, we evaluate models across naturalness (mean opinion and comparative mean opinion score), intelligibility (word error rate), and speaker consistency. SBV2JE matches human ground truth in naturalness (MOS 4.37 vs. 4.38), achieves lower WER, and shows slight preference in CMOS. Enhanced by pitch-accent controls and a WavLM-based discriminator, SBV2JE proves effective for applications like language learning and character dialogue generation, despite higher computational demands. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17320 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Comparative Evaluation of Expressive Japanese Character Text-to-Speech with VITS and Style-BERT-VITS2 Rackauckas, Zackary Hirschberg, Julia Computation and Language Sound Audio and Speech Processing Synthesizing expressive Japanese character speech poses unique challenges due to pitch-accent sensitivity and stylistic variability. This paper empirically evaluates two open-source text-to-speech models--VITS and Style-BERT-VITS2 JP Extra (SBV2JE)--on in-domain, character-driven Japanese speech. Using three character-specific datasets, we evaluate models across naturalness (mean opinion and comparative mean opinion score), intelligibility (word error rate), and speaker consistency. SBV2JE matches human ground truth in naturalness (MOS 4.37 vs. 4.38), achieves lower WER, and shows slight preference in CMOS. Enhanced by pitch-accent controls and a WavLM-based discriminator, SBV2JE proves effective for applications like language learning and character dialogue generation, despite higher computational demands. |
| title | Comparative Evaluation of Expressive Japanese Character Text-to-Speech with VITS and Style-BERT-VITS2 |
| topic | Computation and Language Sound Audio and Speech Processing |
| url | https://arxiv.org/abs/2505.17320 |