Source note2 min read
VoxENES 2026: Benchmarking Generalization of Speech Spoofing Detectors Against LLM-Era TTS and Voice Conversion
Source
arXiv
arxiv.org/abs/2607.11706v1
Source
- Source name: arXiv
- Source kind: paper
- Authors / organization: Aastha Sharma, Guangjing Wang
- Venue: arXiv: cs.SD
- Published: 2026-07-13
- Collected: 2026-07-14
- URL: http://arxiv.org/abs/2607.11706v1
- PDF: https://arxiv.org/pdf/2607.11706v1
- DOI: UNKNOWN
Plain-Language Brief
The paper introduces VoxENES 2026, a new benchmark for testing speech synthesis systems against modern text-to-speech (TTS) and voice conversion (VC) methods. The benchmark includes 53,628 audio samples in English and Spanish, generated using 10 contemporary synthesis methods. When tested with eight pretrained detectors, the benchmark reveals substantial performance degradation, particularly for newer TTS and VC methods.
What the Document Says
- Modern LLM-driven text-to-speech (TTS) and voice conversion (VC) systems differ from those in legacy spoofing benchmarks, creating a generalization gap.
- VoxENES 2026 is a new bilingual benchmark with 53,628 audio samples generated using 10 contemporary speech synthesis methods.
- VoxENES 2026 includes 10 standardized post-processing conditions.
- Eight pretrained detectors were tested on VoxENES 2026 without fine-tuning.
- The best detector achieved an Equal Error Rate (EER) of 28.98% overall, while most performed near or below random chance with modern generators and perturbations.
- Results suggest that current detectors rely on brittle artifacts and VoxENES 2026 can help develop robust audio spoofing countermeasures.
Key Concepts Mentioned
- Speech synthesis
- Text-to-speech (TTS)
- Voice conversion (VC)
- Spoofing detectors
- Generalization gap
- Benchmark
- Audio samples
- Equal Error Rate (EER)
Claims Made by the Document
- Modern TTS and VC systems differ from legacy benchmarks.
- VoxENES 2026 bridges the generalization gap.
- Current detectors rely on brittle artifacts.
Evidence Used
- Mention of modern LLM-driven TTS and VC systems.
- Description of VoxENES 2026 as a new benchmark.
- Results showing performance degradation of detectors with modern generators.
Methods / Approach
UNKNOWN
Limitations Stated by the Authors
UNKNOWN
Limitations Noted by the Agent
Agent interpretation — not from the source.
UNKNOWN
Open Questions
UNKNOWN
Notes for Future Q&A
May include interpretation. Labeled where it does.
UNKNOWN
Raw Abstract / Excerpt
Modern LLM-driven text-to-speech (TTS) and voice conversion (VC) systems produce synthetic speech that differs from the generators represented in many legacy spoofing benchmarks. This mismatch creates a temporal generalization gap that can overestimate detector robustness under real-world post-processing conditions. We bridge this gap by introducing VoxENES 2026, a bilingual (English and Spanish) benchmark of 53,628 audio samples generated using 10 contemporary speech synthesis methods and evaluated under 10 standardized post-processing conditions. Using VoxENES 2026, we benchmark eight pretrained detectors without fine-tuning and observe substantial performance degradation: the best model achieves 28.98% EER overall, while most perform near or below random chance across modern generators and perturbations. Our results highlight the reliance on brittle artifacts in current detectors and establish VoxENES 2026 as a practical testbed for developing robust audio spoofing countermeasures.
