DNSMOS

DNSMOS is a dataset that contains perceptual mean opinion scores (MOS) for evaluating speech quality, used as pseudo-labels in the context of speech quality assessment.

Papers using DNSMOS (8)

CMGAN: Conformer-based Metric-gan For Monaural Speech Enhancement2022 · 93 citesA Study On Zero-shot Non-intrusive Speech Assessment Using Large Language Models2024 · 5 citesReducing The Gap Between Pretrained Speech Enhancement And Recognition Models Using A Real Speech-trained Bridging Module2025 · 1 citesFew-Shot and Pseudo-Label Guided Speech Quality Evaluation with Large Language Models2026When Fine-Tuning Fails and when it Generalises: Role of Data Diversity and Mixed Training in LLM-based TTS2026Reducing the Gap Between Pretrained Speech Enhancement and Recognition Models Using a Real Speech-Trained Bridging Module2025A Study on Zero-shot Non-intrusive Speech Assessment using Large Language Models2024Distil-DCCRN: A Small-footprint DCCRN Leveraging Feature-based Knowledge Distillation in Speech Enhancement2024