AudioInceptionNeXt：TCL AI LAB 对 2023 年 EPIC-SOUND 基于音频的交互识别挑战赛的提交
AudioInceptionNeXt: TCL AI LAB Submission to EPIC-SOUND Audio-Based-Interaction-Recognition Challenge 2023
AI总结：
提出单流 CNN 架构 AudioInceptionNeXt，通过并行多尺度深度可分离卷积核处理对数梅尔谱图，在 EPIC-SOUNDS 2023 测试集上取得 55.43% top-1 准确率并位列第一。
摘要：
本报告介绍了我们提交参加 2023 年 Epic-Kitchen EPIC-SOUNDS 基于音频的交互识别挑战赛的技术细节。该任务旨在学习从音频样本到其对应动作标签的映射。为实现这一目标，我们提出了一种简单而有效的、基于 CNN 的单流架构 AudioInceptionNeXt，该架构作用于音频样本的时频对数梅尔谱图。受 InceptionNeXt 设计的启发，我们在 AudioInceptionNeXt 模块中提出了并行的多尺度深度可分离卷积核，使模型能够更有效地学习时间和频率信息。大尺度可分离核捕捉活动的长持续时间和全局频率语义信息，而小尺度可分离核捕捉活动的短持续时间和频率信息的局部细节。我们的方法在挑战赛测试集上取得了 55.43% 的 top-1 准确率，在公开排行榜上排名第一。代码可在 https://github.com/StevenLauHKHK/AudioInceptionNeXt.git 匿名获取。
英文摘要：
This report presents the technical details of our submission to the 2023 Epic-Kitchen EPIC-SOUNDS Audio-Based Interaction Recognition Challenge. The task is to learn the mapping from audio samples to their corresponding action labels. To achieve this goal, we propose a simple yet effective single-stream CNN-based architecture called AudioInceptionNeXt that operates on the time-frequency log-mel-spectrogram of the audio samples. Motivated by the design of the InceptionNeXt, we propose parallel multi-scale depthwise separable convolutional kernels in the AudioInceptionNeXt block, which enable the model to learn the time and frequency information more effectively. The large-scale separable kernels capture the long duration of activities and the global frequency semantic information, while the small-scale separable kernels capture the short duration of activities and local details of frequency information. Our approach achieved 55.43% of top-1 accuracy on the challenge test set, ranked as 1st on the public leaderboard. Codes are available anonymously at https://github.com/StevenLauHKHK/AudioInceptionNeXt.git.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
