analiza aiqbench
This research introduces NVMOS, the first model capable of reliably predicting perceptual quality for non-verbal vocalizations like laughter and sighs in speech synthesis. By demonstrating that general-purpose multimodal models fail to accurately assess these specific acoustic cues compared to human experts, the study highlights a critical gap in current AI evaluation metrics. The proposed solution utilizes a specialized local focusing module to achieve expert-level agreement, advancing the precision of audio quality assessment for emotional intent and naturalness.
NVMOS: Non-Verbal Vocalization Quality Assessment in Speech
arXiv:2606.15888v1 Announce Type: cross Abstract: Non-verbal vocalizations (NVs), such as laughter, sighs, and coughs, are important acoustic cues for emotion and intent. Existing speech quality assessment methods typically focus on overall naturalness, while non-verbal TTS evaluations mainly examine whether a target NV appears with the correct type and position. However, the perceptual quality of NV events themselves remains underexplored. To address this gap, we construct an NV-MOS dataset containing outputs from multiple NV-TTS systems and naturally occurring NV samples, with ratings collected from three acoustic experts on a perceptual quality scale. We further analyze audio-capable multimodal large language models such as Gemini and find clear inconsistencies between their scores and expert ratings. These results suggest that general-purpose multimodal models cannot reliably replace human judgments for NV quality assessment. We then propose NVMOS, to our knowledge the first model that can reliably predict the perceptual quality of NV events in speech. Experimental results show that, with a local NV-event focusing module, NVMOS reaches expert-level or stronger agreement with human MOS.
Abstract:Non-verbal vocalizations (NVs), such as laughter, sighs, and coughs, are important acoustic cues for emotion and intent. Existing speech quality assessment methods typically focus on overall naturalness, while non-verbal TTS evaluations mainly examine whether a target NV appears with the correct type and position. However, the perceptual quality of NV events themselves remains underexplored. To address this gap, we construct an NV-MOS dataset containing outputs from multiple NV-TTS systems and naturally occurring NV samples, with ratings collected from three acoustic experts on a perceptual quality scale. We further analyze audio-capable multimodal large language models such as Gemini and find clear inconsistencies between their scores and expert ratings. These results suggest that general-purpose multimodal models cannot reliably replace human judgments for NV quality assessment. We then propose NVMOS, to our knowledge the first model that can reliably predict the perceptual quality of NV events in speech. Experimental results show that, with a local NV-event focusing module, NVMOS reaches expert-level or stronger agreement with human MOS.
6 pages. Code and model: this https URL
Sound (cs.SD); Artificial Intelligence (cs.AI); Audio and Speech Processing (eess.AS)
arXiv:2606.15888 [cs.SD]
(or arXiv:2606.15888v1 [cs.SD] for this version)
https://doi.org/10.48550/arXiv.2606.15888 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 273 cuvinte
Subiecte conexe
Interpretable Singing Emotion Recognition Using Acoustic and Vocal Technique Features
arXiv · 27 oct. 2026
WoW-Bench: Evaluating Fine-Grained Acoustic Perception in Audio-Language Models via Marine Mammal Vocalizations
arXiv · 21 oct. 2026
Uncertainty-Aware Crossmodal Fusion for Classification of Animal Behavior
arXiv · 5 aug. 2026
A Mid - 2026 Primer On Cybersecurity And Addressing New Threats
forbes.com · 31 iul. 2026