Complex-Text Robustness Evaluation and Failure Diagnosis for Low-Resource Multilingual Text-to-Speech

• URL: http://arxiv.org/abs/2609.11545v1
• Date: Thu, 10 Sep 2026 13:41:22 GMT
• Title: Complex-Text Robustness Evaluation and Failure Diagnosis for Low-Resource Multilingual Text-to-Speech
• Abstract summary: Low-resource multilingual text-to-speech (TTS) systems have expanded language coverage, but their robustness under complex text inputs remains insufficiently diagnosed.<n>This paper proposes a complex-text robustness diagnosis framework for low-resource multilingual TTS.<n>A multilingual robustness testing scheme is designed for Thai, Vietnamese, Swahili, and Indonesian, covering ordinary sentences and multiple types of complex text inputs.
• Score: 10.008502036976243
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Low-resource multilingual text-to-speech (TTS) systems have expanded language coverage, but their robustness under complex text inputs remains insufficiently diagnosed. Existing evaluations mainly focus on naturalness, speaker similarity, and content consistency using regular test sentences, while providing limited insight into how multilingual TTS systems fail when handling challenging inputs such as numbers, dates, named entities, long sentences, code-switched expressions, and punctuation-related structures. This paper proposes a complex-text robustness diagnosis framework for low-resource multilingual TTS. We evaluate robustness from three dimensions: content consistency, language consistency, and generation stability. A multilingual robustness testing scheme is designed for Thai, Vietnamese, Swahili, and Indonesian, covering ordinary sentences and multiple types of complex text inputs. We further introduce automatic diagnostic metrics, including character error rate, language identification accuracy, and duration abnormal rate. To support input-level risk analysis before speech generation, we propose a lightweight Text Risk Score (TRS), which estimates synthesis risk from interpretable text features without manual annotation or model training. Experiments on three representative multilingual TTS systems, including OmniVoice, VoxCPM2, and MMS-TTS, show that complex text inputs expose systematic failure patterns that are not fully reflected by ordinary short-sentence evaluation. Different systems exhibit distinct vulnerabilities in number normalization, named entity handling, long-text generation, and code-switched input processing. Furthermore, TRS shows a positive correlation with content errors and duration abnormalities, demonstrating its usefulness as a low-cost pre-synthesis indicator for complex-text risk diagnosis in low-resource multilingual TTS.

Related papers

• IndicDetect: Evaluating Cross-Lingual LLM-Generated Text Detection for Hindi, Telugu, and Tamil [32.21285778019889]
  We present a benchmark for AI-generated text detection in Hindi, Telugu, and Tamil.<n>IndicDetect is designed to assess the robustness of detectors under realistic distribution shifts.
  arXiv  Detail & Related papers  (2026-08-30T17:21:55Z)
• SFMS-ALR: Script-First Multilingual Speech Synthesis with Adaptive Locale Resolution [0.0]
  Intra-sentence multilingual speech synthesis (code-switching TTS) remains a major challenge due to abrupt language shifts, varied scripts, and mismatched prosody between languages.<n>We introduce Script-First Multilingual Synthesis with Adaptive Locale Resolution (SFMS-ALR), an engine-agnostic framework for fluent, real-time code-switched speech generation.
  arXiv  Detail & Related papers  (2025-10-27T21:39:07Z)
• Human Texts Are Outliers: Detecting LLM-generated Texts via Out-of-distribution Detection [71.59834293521074]
  We develop a framework to distinguish between human-authored and machine-generated text.<n>Our method achieves 98.3% AUROC and AUPR with only 8.9% FPR95 on DeepFake dataset.<n>Code, pretrained weights, and demo will be released.
  arXiv  Detail & Related papers  (2025-10-07T08:14:45Z)
• Enhancing Robustness of Autoregressive Language Models against Orthographic Attacks via Pixel-based Approach [51.95266411355865]
  Autoregressive language models are vulnerable to orthographic attacks.<n>This vulnerability stems from the out-of-vocabulary issue inherent in subword tokenizers and their embeddings.<n>We propose a pixel-based generative language model that replaces the text-based embeddings with pixel-based representations by rendering words as individual images.
  arXiv  Detail & Related papers  (2025-08-28T20:48:38Z)
• PromotionGo at SemEval-2025 Task 11: A Feature-Centric Framework for Cross-Lingual Multi-Emotion Detection in Short Texts [1.210852962855694]
  This paper presents our system for SemEval 2025 Task 11: Bridging the Gap in Text-Based Emotion Detection.<n>We propose a feature-centric framework that dynamically adapts document representations and learning algorithms to optimize language-specific performance.
  arXiv  Detail & Related papers  (2025-07-11T11:21:18Z)
• Human-in-the-Loop Generation of Adversarial Texts: A Case Study on Tibetan Script [12.108238610034961]
  We introduce HITL-GAT, an interactive system based on a general approach to human-in-the-loop generation of adversarial texts.<n>We demonstrate the utility of HITL-GAT through a case study on Tibetan script, employing three customized adversarial text generation methods.
  arXiv  Detail & Related papers  (2024-12-17T02:29:54Z)
• Vulnerability of LLMs to Vertically Aligned Text Manipulations [130.54118945532898]
  Vertical text input is commonly encountered in various real-world applications, such as mathematical computations and word-based Sudoku puzzles.<n>Recent research demonstrates that modifying input formats, such as vertically aligning words for encoder-based models, can substantially lower accuracy in text classification tasks.
  arXiv  Detail & Related papers  (2024-10-26T00:16:08Z)
• Decomposed Prompting: Probing Multilingual Linguistic Structure Knowledge in Large Language Models [54.58989938395976]
  We introduce a decomposed prompting approach for sequence labeling tasks.<n>We test our method on the Universal Dependencies part-of-speech tagging dataset for 38 languages.
  arXiv  Detail & Related papers  (2024-02-28T15:15:39Z)
• Not Enough Labeled Data? Just Add Semantics: A Data-Efficient Method for Inferring Online Health Texts [0.0]
  We employ Abstract Representation (AMR) graphs as a means to model low-resource Health NLP tasks. AMRs are well suited to model online health texts as they represent multi-sentence inputs, abstract away from complex terminology, and model long-distance relationships. Our experiments show that we can improve performance on 6 low-resource health NLP tasks by augmenting text embeddings with semantic graph embeddings.
  arXiv  Detail & Related papers  (2023-09-18T15:37:30Z)
• Code-Switching Text Generation and Injection in Mandarin-English ASR [57.57570417273262]
  We investigate text generation and injection for improving the performance of an industry commonly-used streaming model, Transformer-Transducer (T-T) We first propose a strategy to generate code-switching text data and then investigate injecting generated text into T-T model explicitly by Text-To-Speech (TTS) conversion or implicitly by tying speech and text latent spaces. Experimental results on the T-T model trained with a dataset containing 1,800 hours of real Mandarin-English code-switched speech show that our approaches to inject generated code-switching text significantly boost the performance of T-T models.
  arXiv  Detail & Related papers  (2023-03-20T09:13:27Z)
• Evaluating the Morphosyntactic Well-formedness of Generated Texts [88.20502652494521]
  We propose L'AMBRE -- a metric to evaluate the morphosyntactic well-formedness of text. We show the effectiveness of our metric on the task of machine translation through a diachronic study of systems translating into morphologically-rich languages.
  arXiv  Detail & Related papers  (2021-03-30T18:02:58Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.