Saved in:
Bibliographic Details
Main Authors: Li, Zhipeng, Xing, Xiaofen, Xing, Jingyuan, Hu, Hangrui, Lu, Heng, Xu, Xiangmin
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2508.14713
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918127873294336
author Li, Zhipeng
Xing, Xiaofen
Xing, Jingyuan
Hu, Hangrui
Lu, Heng
Xu, Xiangmin
author_facet Li, Zhipeng
Xing, Xiaofen
Xing, Jingyuan
Hu, Hangrui
Lu, Heng
Xu, Xiangmin
contents In long-text speech synthesis, current approaches typically convert text to speech at the sentence-level and concatenate the results to form pseudo-paragraph-level speech. These methods overlook the contextual coherence of paragraphs, leading to reduced naturalness and inconsistencies in style and timbre across the long-form speech. To address these issues, we propose a Context-Aware Memory (CAM)-based long-context Text-to-Speech (TTS) model. The CAM block integrates and retrieves both long-term memory and local context details, enabling dynamic memory updates and transfers within long paragraphs to guide sentence-level speech synthesis. Furthermore, the prefix mask enhances the in-context learning ability by enabling bidirectional attention on prefix tokens while maintaining unidirectional generation. Experimental results demonstrate that the proposed method outperforms baseline and state-of-the-art long-context methods in terms of prosody expressiveness, coherence and context inference cost across paragraph-level speech.
format Preprint
id arxiv_https___arxiv_org_abs_2508_14713
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Long-Context Speech Synthesis with Context-Aware Memory
Li, Zhipeng
Xing, Xiaofen
Xing, Jingyuan
Hu, Hangrui
Lu, Heng
Xu, Xiangmin
Audio and Speech Processing
Sound
In long-text speech synthesis, current approaches typically convert text to speech at the sentence-level and concatenate the results to form pseudo-paragraph-level speech. These methods overlook the contextual coherence of paragraphs, leading to reduced naturalness and inconsistencies in style and timbre across the long-form speech. To address these issues, we propose a Context-Aware Memory (CAM)-based long-context Text-to-Speech (TTS) model. The CAM block integrates and retrieves both long-term memory and local context details, enabling dynamic memory updates and transfers within long paragraphs to guide sentence-level speech synthesis. Furthermore, the prefix mask enhances the in-context learning ability by enabling bidirectional attention on prefix tokens while maintaining unidirectional generation. Experimental results demonstrate that the proposed method outperforms baseline and state-of-the-art long-context methods in terms of prosody expressiveness, coherence and context inference cost across paragraph-level speech.
title Long-Context Speech Synthesis with Context-Aware Memory
topic Audio and Speech Processing
Sound
url https://arxiv.org/abs/2508.14713