paper-with-me

Papers

ZeSTA: Zero-Shot TTS Augmentation with Domain-Conditioned Training for Data-Efficient Personalized Speech Synthesis

2026-03-04 · Youngwon Choi, Jinwoo Oh, Hwayeon Kim, Hyeonyu Kim arxiv

We investigate the use of zero-shot text-to-speech (ZS-TTS) as a data augmentation source for low-resource personalized speech synthesis. While synthetic augmentation can provide linguistically rich and phonetically diverse speech, naively mixing large amounts of synthetic speech with limited real recordings often leads to speaker similarity degradation during fine-tuning. To address this issue, we propose ZeSTA, a simple domain-conditioned training framework that distinguishes real and synthetic speech via a lightweight domain embedding, combined with real-data oversampling to stabilize adaptation under extremely limited target data, without modifying the base architecture. Experiments on LibriTTS and an in-house dataset with two ZS-TTS sources demonstrate that our approach improves speaker similarity over naive synthetic augmentation while preserving intelligibility and perceptual quality. Audio samples are available on our web page.

📄 PDF Abstract BibTeX arXiv:2603.04219

Code (0)

등록된 구현이 없습니다.

Tasks

Data AugmentationSpeech Synthesis

Similar Papers 제목 키워드 기반

Accelerating New Product Introduction for Visual Quality Inspection via Few-Shot Diffusion-Based Defect Synthesis

2026-04-22 · Serkan Hamdi Güğül, Kemal Levi, Burak Acar arxiv

Industrial visual inspection systems often suffer from a severe scarcity of labeled defect data, particularly during the early stages of New Product Introduction (NPI). This limitation hinders the deployment of robust su…

Representation LearningData AugmentationDomain Adaptation

Schema Augmentation for Zero-Shot Domain Adaptation in Dialogue State Tracking

2024-10-31 · Christopher Richardson, Roshan Sharma, Neeraj Gaur, Parisa Haghani 외

Zero-shot domain adaptation for dialogue state tracking (DST) remains a challenging problem in task-oriented dialogue (TOD) systems, where models must generalize to target domains unseen at training time. Current large l…

Data AugmentationDialogue State TrackingDomain AdaptationLanguage Modeling+3

MALM: Mixing Augmented Language Modeling for Zero-Shot Machine Translation

2022-10-01 · Kshitij Gupta

Large pre-trained language models have brought remarkable progress in NLP. Pre-training and Fine-tuning have given state-of-art performance across tasks in text processing. Data Augmentation techniques have also helped b…

Data AugmentationLanguage ModelingLanguage ModellingMachine Translation+2

Generate then Refine: Data Augmentation for Zero-shot Intent Detection

2024-10-02 · I-Fan Lin, Faegheh Hasibi, Suzan Verberne

In this short paper we propose a data augmentation method for intent detection in zero-resource domains. Existing data augmentation methods rely on few labelled examples for each intent category, which can be expensive i…

Data AugmentationDiversityIntent DetectionLanguage Modeling+2

Large Language Models for Document-Level Event-Argument Data Augmentation for Challenging Role Types

2024-03-05 · Joseph Gatto, Parker Seegmiller, Omar Sharif, Sarah M. Preum

Event Argument Extraction (EAE) is an extremely difficult information extraction problem -- with significant limitations in few-shot cross-domain (FSCD) settings. A common solution to FSCD modeling is data augmentation. …

Data AugmentationEvent Argument Extraction