# TTSlow: Slow Down Text-to-Speech with Efficiency Robustness Evaluations

arXiv: https://arxiv.org/abs/2407.01927 · HTML: https://mlsanity.com/paper/2407.01927/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 2 Jul 2024.

## Built on by (most cited first, 13 of 13)

- [VoiceBench: Benchmarking LLM-Based Voice Assistants](https://mlsanity.com/paper/2410.17196/) (2024)
- [Safety at Scale: A Comprehensive Survey of Large Model and Agent Safety](https://mlsanity.com/paper/2502.05206/) (2025)
- [CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks](https://mlsanity.com/paper/2507.10535/) (2025)
- [Benchmarking and Confidence Evaluation of LALMs For Temporal Reasoning](https://mlsanity.com/paper/2505.13115/) (2025)
- [RemedyGS: Defend 3D Gaussian Splatting against Computation Cost Attacks](https://mlsanity.com/paper/2511.22147/) (2025)
- [Beyond Single-Audio: Advancing Multi-Audio Processing in Audio Large Language Models](https://mlsanity.com/paper/2409.18680/) (2024)
- [Towards Explicit Acoustic Evidence Perception in Audio LLMs for Speech Deepfake Detection](https://mlsanity.com/paper/2601.23066/) (2026)
- [Exploring Gradient-Guided Masked Language Model to Detect Textual Adversarial Attacks](https://mlsanity.com/paper/2504.08798/) (2025)
- [Prompt-Unseen-Emotion: Zero-shot Expressive Speech Synthesis with Prompt-LLM Contextual Knowledge for Mixed Emotions](https://mlsanity.com/paper/2506.02742/) (2025)
- [Efficiency Robustness of Dynamic Deep Learning Systems](https://mlsanity.com/paper/2506.10831/) (2025)
- [MultiGen: Child-Friendly Multilingual Speech Generator with LLMs](https://mlsanity.com/paper/2508.08715/) (2025)
- [PIMbot: A Self-Adaptive Attack Framework for Adversarial Manipulation of Multi-Robot Reinforcement Learning](https://mlsanity.com/paper/2605.23027/) (2026)
- [MORE: Multi-Objective Adversarial Attacks on Speech Recognition](https://mlsanity.com/paper/2601.01852/) (2026)

## Abstract (arXiv)

Text-to-speech (TTS) has been extensively studied for generating high-quality speech with textual inputs, playing a crucial role in various real-time applications. For real-world deployment, ensuring stable and timely generation in TTS models against minor input perturbations is of paramount importance. Therefore, evaluating the robustness of TTS models against such perturbations, commonly known as adversarial attacks, is highly desirable. In this paper, we propose TTSlow, a novel adversarial approach specifically tailored to slow down the speech generation process in TTS systems. To induce long TTS waiting time, we design novel efficiency-oriented adversarial loss to encourage endless generation process. TTSlow encompasses two attack strategies targeting both text inputs and speaker embedding. Specifically, we propose TTSlow-text, which utilizes a combination of homoglyphs-based and swap-based perturbations, along with TTSlow-spk, which employs a gradient optimization attack approach for speaker embedding. TTSlow serves as the first attack approach targeting a wide range of TTS models, including autoregressive and non-autoregressive TTS ones, thereby advancing exploration in audio security. Extensive experiments are conducted to evaluate the inference efficiency of TTS models, and in-depth analysis of generated speech intelligibility is performed using Gemini. The results demonstrate that TTSlow can effectively slow down two TTS models across three publicly available datasets. We are committed to releasing the source code upon acceptance, facilitating further research and benchmarking in this domain.