paper-with-me

홈 › Papers

DRASP: A Dual-Resolution Attentive Statistics Pooling Framework for Automatic MOS Prediction

2025-08-29 · Cheng-Yeh Yang, Kuan-Tang Huang, Chien-Chun Wang, Hung-Shin Lee, Hsin-Min Wang, Berlin Chen arxiv

A pooling mechanism is essential for mean opinion score (MOS) prediction, facilitating the transformation of variable-length audio features into a concise fixed-size representation that effectively encodes speech quality. Existing pooling methods typically operate at a singular granularity, concentrating either on a comprehensive global perspective or a detailed frame-level analysis, which may overlook complementary perceptual insights. To address this limitation, we introduce the Dual-Resolution Attentive Statistics Pooling (DRASP) framework. DRASP integrates both coarse-grained, global statistical summaries and fine-grained, attentive analyses of perceptually significant segments. This dual-view architecture empowers our model to formulate a more thorough and robust representation, capturing both the overarching structural context and salient local details concurrently. Extensive experiments validate the effectiveness and strong generalization ability of the proposed framework. It consistently outperforms various baseline methods across diverse datasets (MusicEval and AES-Natural), MOS prediction backbones (including a CLAP-based model and AudioBox-Aesthetics), and different audio generation systems, achieving a relative improvement of 10.39% in system-level Spearman's rank correlation coefficient (SRCC) over the widely-used average pooling approach.

📄 PDF Abstract BibTeX arXiv:2508.21407

Code (0)

등록된 구현이 없습니다.

Tasks

Audio Generation

Similar Papers 제목 키워드 기반

Attentive pooling for Group Activity Recognition

2022-08-31 · Ding Li, Yuan Xie, Wensheng Zhang, Yongqiang Tang 외

In group activity recognition, hierarchical framework is widely adopted to represent the relationships between individuals and their corresponding group, and has achieved promising performance. However, the existing meth…

Activity RecognitionGroup Activity Recognition

Attentive Statistics Pooling for Deep Speaker Embedding

2019-02-25

This paper proposes attentive statistics pooling for deep speaker embedding in text-independent speaker verification. In conventional speaker embedding, frame-level features are averaged over all the frames of a single u…

Speaker VerificationText-Independent Speaker Verification

SoCov: Semi-Orthogonal Parametric Pooling of Covariance Matrix for Speaker Recognition

2025-04-23 · Rongjin Li, Weibin Zhang, Dongpeng Chen, Jintao Kang 외

In conventional deep speaker embedding frameworks, the pooling layer aggregates all frame-level features over time and computes their mean and standard deviation statistics as inputs to subsequent segment-level layers. S…

Speaker Recognition

Learning Audio-Visual embedding for Person Verification in the Wild

2022-09-09 · Peiwen Sun, Shanshan Zhang, Zishan Liu, Yougen Yuan 외

It has already been observed that audio-visual embedding is more robust than uni-modality embedding for person verification. Here, we proposed a novel audio-visual strategy that considers aggregators from a fusion perspe…

Face Verification

Transport-Oriented Feature Aggregation for Speaker Embedding Learning

2022-06-26 · Yusheng Tian, Jingyu Li, Tan Lee

Pooling is needed to aggregate frame-level features into utterance-level representations for speaker modeling. Given the success of statistics-based pooling methods, we hypothesize that speaker characteristics are well r…

Speaker Verification