GatherMOS uses LLMs as meta-evaluators to aggregate acoustic features and pseudo-labels for improved mean opinion score prediction in few-shot speech quality assessment.
Zero-shot GatherMOS Given an input waveformx∈R T , we extract several acous- tic descriptors that summarize temporal, spectral, and percep- tual information
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Few-Shot and Pseudo-Label Guided Speech Quality Evaluation with Large Language Models
GatherMOS uses LLMs as meta-evaluators to aggregate acoustic features and pseudo-labels for improved mean opinion score prediction in few-shot speech quality assessment.