# Comparative Evaluation of Expressive Japanese Character Text-to-Speech with VITS and Style-BERT-VITS2

arXiv: https://arxiv.org/abs/2505.17320 · HTML: https://mlsanity.com/paper/2505.17320/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 14 of its 28 references that are in the map.
- First posted on arXiv 22 May 2025.

## Builds on (most cited first, 14 of 14)

- [FastSpeech 2: Fast and High-Quality End-to-End Text to Speech](https://mlsanity.com/paper/2006.04558/) (2020)
- [Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech](https://mlsanity.com/paper/2106.06103/) (2021)
- [Tacotron: Towards End-to-End Speech Synthesis](https://mlsanity.com/paper/1703.10135/) (2017)
- [Wespeaker: A Research and Production oriented Speaker Embedding Learning Toolkit](https://mlsanity.com/paper/2210.17016/) (2022)
- [PnG BERT: Augmented BERT on Phonemes and Graphemes for Neural TTS](https://mlsanity.com/paper/2103.15060/) (2021)
- [VITS2: Improving Quality and Efficiency of Single-Stage Text-to-Speech with Adversarial Learning and Architecture Design](https://mlsanity.com/paper/2307.16430/) (2023)
- [End-to-End Text-to-Speech using Latent Duration based on VQ-VAE](https://mlsanity.com/paper/2010.09602/) (2020)
- [Learning Speaker Embedding from Text-to-Speech](https://mlsanity.com/paper/2010.11221/) (2020)
- [Improving Speech Prosody of Audiobook Text-to-Speech Synthesis with Acoustic and Textual Contexts](https://mlsanity.com/paper/2211.02336/) (2022)
- [Polyphone disambiguation and accent prediction using pre-trained language models in Japanese TTS front-end](https://mlsanity.com/paper/2201.09427/) (2022)
- [Speaker consistency loss and step-wise optimization for semi-supervised joint training of TTS and ASR using unpaired text data](https://mlsanity.com/paper/2207.04659/) (2022)
- [Stable-TTS: Stable Speaker-Adaptive Text-to-Speech Synthesis via Prosody Prompting](https://mlsanity.com/paper/2412.20155/) (2024)
- [Adversarial Speaker-Consistency Learning Using Untranscribed Speech Data for Zero-Shot Multi-Speaker Text-to-Speech](https://mlsanity.com/paper/2210.05979/) (2022)
- [Investigation of Japanese PnG BERT language model in text-to-speech synthesis for pitch accent language](https://mlsanity.com/paper/2212.08321/) (2022)

## Built on by (most cited first, 1 of 1)

- [Re:Member: Emotional Question Generation from Personal Memories](https://mlsanity.com/paper/2510.19030/) (2025)

## Abstract (arXiv)

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.