# Phrase break prediction with bidirectional encoder representations in Japanese text-to-speech synthesis

arXiv: https://arxiv.org/abs/2104.12395 · HTML: https://mlsanity.com/paper/2104.12395/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 29 references that are in the map.
- First posted on arXiv 26 Apr 2021.

## Builds on (most cited first, 11 of 11)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [A Primer in BERTology: What we know about how BERT works](https://mlsanity.com/paper/2002.12327/) (2020)
- [Natural TTS Synthesis by Conditioning WaveNet on Mel Spectrogram Predictions](https://mlsanity.com/paper/1712.05884/) (2017)
- [FastSpeech 2: Fast and High-Quality End-to-End Text to Speech](https://mlsanity.com/paper/2006.04558/) (2020)
- [Tacotron: Towards End-to-End Speech Synthesis](https://mlsanity.com/paper/1703.10135/) (2017)
- [Parallel WaveGAN: A fast waveform generation model based on generative adversarial networks with multi-resolution spectrogram](https://mlsanity.com/paper/1910.11480/) (2019)
- [Transformer based Grapheme-to-Phoneme Conversion](https://mlsanity.com/paper/2004.06338/) (2020)
- [Representation Mixing for TTS Synthesis](https://mlsanity.com/paper/1811.07240/) (2018)
- [Parallel waveform synthesis based on generative adversarial networks with voicing-aware conditional discriminators](https://mlsanity.com/paper/2010.14151/) (2020)

## Built on by (most cited first, 12 of 12)

- [Duration-aware pause insertion using pre-trained language model for multi-speaker text-to-speech](https://mlsanity.com/paper/2302.13652/) (2023)
- [Assessing Phrase Break of ESL Speech with Pre-trained Language Models and Large Language Models](https://mlsanity.com/paper/2306.04980/) (2023)
- [PauseSpeech: Natural Speech Synthesis via Pre-trained Language Model and Pause-based Prosody Modeling](https://mlsanity.com/paper/2306.07489/) (2023)
- [An investigation of phrase break prediction in an End-to-End TTS system](https://mlsanity.com/paper/2304.04157/) (2023)
- [Cross-Lingual Transfer Learning for Phrase Break Prediction with Multilingual Language Model](https://mlsanity.com/paper/2306.02579/) (2023)
- [ProsodyFM: Unsupervised Phrasing and Intonation Control for Intelligible Speech Synthesis](https://mlsanity.com/paper/2412.11795/) (2024)
- [Synthetic Data Generation for Phrase Break Prediction with Large Language Model](https://mlsanity.com/paper/2507.18044/) (2025)
- [Cross-Utterance Conditioned VAE for Non-Autoregressive Text-to-Speech](https://mlsanity.com/paper/2205.04120/) (2022)
- [Cross-Utterance Conditioned VAE for Speech Generation](https://mlsanity.com/paper/2309.04156/) (2023)
- [Improving French Synthetic Speech Quality via SSML Prosody Control](https://mlsanity.com/paper/2508.17494/) (2025)
- [Assessing Phrase Break of ESL speech with Pre-trained Language Models](https://mlsanity.com/paper/2210.16029/) (2022)
- [LLM-Based Multi-Reference Evaluation for Efficient and Robust Assessment of Phrase Break Annotations](https://mlsanity.com/paper/2606.21098/) (2026)

## Abstract (arXiv)

We propose a novel phrase break prediction method that combines implicit features extracted from a pre-trained large language model, a.k.a BERT, and explicit features extracted from BiLSTM with linguistic features. In conventional BiLSTM based methods, word representations and/or sentence representations are used as independent components. The proposed method takes account of both representations to extract the latent semantics, which cannot be captured by previous methods. The objective evaluation results show that the proposed method obtains an absolute improvement of 3.2 points for the F1 score compared with BiLSTM-based conventional methods using linguistic features. Moreover, the perceptual listening test results verify that a TTS system that applied our proposed method achieved a mean opinion score of 4.39 in prosody naturalness, which is highly competitive with the score of 4.37 for synthesized speech with ground-truth phrase breaks.