被撤回的 arXiv 论文：Adaptive Duration Model for Text Speech Alignment

This paper has been withdrawn by Junjie Cao

No PDF available, click to view other formats

Abstract:Speech-to-text alignment is a critical component of neural text to speech (TTS) models. Autoregressive TTS models typically use an attention mechanism to learn these alignments on-line, while non-autoregressive end to end TTS models rely on durations extracted from external sources. In this paper, we propose a novel duration prediction framework that can give promising phoneme-level duration distribution with given text. In our experiments, the proposed duration model has more precise prediction and adaptation ability to conditions, compared to previous baseline models. Specifically, it makes a considerable improvement on phoneme-level alignment accuracy and makes the performance of zero-shot TTS models more robust to the mismatch between prompt audio and input audio.

Comments:	the work is not complete, it is a bad work and outdated
Subjects:	Sound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
Cite as:	arXiv:2507.22612 [cs.SD]
	(or arXiv:2507.22612v3 [cs.SD] for this version)
	https://doi.org/10.48550/arXiv.2507.22612 arXiv-issued DOI via DataCite

Submission history

From: Junjie Cao [view email]
[v1] Wed, 30 Jul 2025 12:31:11 UTC (150 KB)
[v2] Fri, 29 Aug 2025 06:09:29 UTC (144 KB)
[v3] Thu, 15 Oct 2026 12:13:45 UTC (1 KB) (withdrawn)