SOTAVerified

EnCLAP++: Analyzing the EnCLAP Framework for Optimizing Automated Audio Captioning Performance

2024-09-02Code Available2· sign in to hype

Jaeyeon Kim, Minjeon Jeon, JaeYoon Jung, Sang Hoon Woo, Jinjoo Lee

Code Available — Be the first to reproduce this paper.

Reproduce

Code

Abstract

In this work, we aim to analyze and optimize the EnCLAP framework, a state-of-the-art model in automated audio captioning. We investigate the impact of modifying the acoustic encoder components, explore pretraining with different dataset scales, and study the effectiveness of a reranking scheme. Through extensive experimentation and quantitative analysis of generated captions, we develop EnCLAP++, an enhanced version that significantly surpasses the original.

Tasks

Benchmark Results

DatasetModelMetricClaimedVerifiedStatus
AudioCapsEnCLAP++-largeSPIDEr0.51Unverified
AudioCapsEnCLAP++-baseSPIDEr0.5Unverified

Reproductions