# Duration-aware pause insertion using pre-trained language model for multi-speaker text-to-speech

## TL;DR

This paper introduces a duration-aware pause insertion method for multi-speaker TTS using a pre-trained BERT model with speaker embeddings, improving naturalness and rhythm in synthetic speech.

## Contribution

It proposes a novel pause insertion framework leveraging BERT and speaker embeddings, specifically addressing multi-speaker variability and duration-aware pause prediction.

## Key findings

- Improved precision and recall in pause prediction
- Enhanced rhythm and naturalness in synthetic speech
- Effective modeling of speaker-specific pause styles

## Abstract

Pause insertion, also known as phrase break prediction and phrasing, is an essential part of TTS systems because proper pauses with natural duration significantly enhance the rhythm and intelligibility of synthetic speech. However, conventional phrasing models ignore various speakers' different styles of inserting silent pauses, which can degrade the performance of the model trained on a multi-speaker speech corpus. To this end, we propose more powerful pause insertion frameworks based on a pre-trained language model. Our approach uses bidirectional encoder representations from transformers (BERT) pre-trained on a large-scale text corpus, injecting speaker embedding to capture various speaker characteristics. We also leverage duration-aware pause insertion for more natural multi-speaker TTS. We develop and evaluate two types of models. The first improves conventional phrasing models on the position prediction of respiratory pauses (RPs), i.e., silent pauses at word transitions without punctuation. It performs speaker-conditioned RP prediction considering contextual information and is used to demonstrate the effect of speaker information on the prediction. The second model is further designed for phoneme-based TTS models and performs duration-aware pause insertion, predicting both RPs and punctuation-indicated pauses (PIPs) that are categorized by duration. The evaluation results show that our models improve the precision and recall of pause insertion and the rhythm of synthetic speech.