paper-with-me

홈 › Papers

An investigation of phrase break prediction in an End-to-End TTS system

2023-04-09 · Anandaswarup Vadapalli

Purpose: This work explores the use of external phrase break prediction models to enhance listener comprehension in End-to-End Text-to-Speech (TTS) systems. Methods: The effectiveness of these models is evaluated based on listener preferences in subjective tests. Two approaches are explored: (1) a bidirectional LSTM model with task-specific embeddings trained from scratch, and (2) a pre-trained BERT model fine-tuned on phrase break prediction. Both models are trained on a multi-speaker English corpus to predict phrase break locations in text. The End-to-End TTS system used comprises a Tacotron2 model with Dynamic Convolutional Attention for mel spectrogram prediction and a WaveRNN vocoder for waveform generation. Results: The listening tests show a clear preference for text synthesized with predicted phrase breaks over text synthesized without them. Conclusion: These results confirm the value of incorporating external phrasing models within End-to-End TTS to enhance listener comprehension.

📄 PDF Abstract BibTeX arXiv:2304.04157

Code (2)

anandaswarup/phrase_break_prediction 공식 구현 pytorch
anandaswarup/rnn-tts 공식 구현 pytorch

Tasks

Predictiontext-to-speechText to Speech

Similar Papers 제목 키워드 기반

Cross-Lingual Transfer Learning for Phrase Break Prediction with Multilingual Language Model

2023-06-05 · Hoyeon Lee, Hyun-Wook Yoon, Jong-Hwan Kim, Jae-Min Kim

Phrase break prediction is a crucial task for improving the prosody naturalness of a text-to-speech (TTS) system. However, most proposed phrase break prediction models are monolingual, trained exclusively on a large amou…

Cross-Lingual TransferLanguage ModelingLanguage ModellingPrediction+3

Synthetic Data Generation for Phrase Break Prediction with Large Language Model

2025-07-24 · Hoyeon Lee, Sejung Son, Ye-Eun Kang, Jong-Hwan Kim arxiv

Current approaches to phrase break prediction address crucial prosodic aspects of text-to-speech systems but heavily rely on vast human annotations from audio or text, incurring significant manual effort and cost. Inhere…

Synthetic Data Generation

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

2021-04-26 · Kosuke Futamata, Byeongseon Park, Ryuichi Yamamoto, Kentaro Tachibana

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 co…

Language ModelingLanguage ModellingLarge Language ModelSentence+4

Open-Source Boundary-Annotated Corpus for Arabic Speech and Language Processing

2012-05-01 · LREC 2012 5 · Claire Brierley, Majdi Sawalha, Eric Atwell

A boundary-annotated and part-of-speech tagged corpus is a prerequisite for developing phrase break classifiers. Boundary annotations in English speech corpora are descriptive, delimiting intonation units perceived by th…

ChunkingDescriptiveSpeech SynthesisText-To-Speech Synthesis

A LSTM Approach with Sub-Word Embeddings for Mongolian Phrase Break Prediction

2018-08-01 · COLING 2018 8 · Rui Liu, Feilong Bao, Guanglai Gao, HUI ZHANG 외

In this paper, we first utilize the word embedding that focuses on sub-word units to the Mongolian Phrase Break (PB) prediction task by using Long-Short-Term-Memory (LSTM) model. Mongolian is an agglutinative language. E…

Dictionary LearningMachine TranslationQuestion AnsweringWord Embeddings