Acted Emotional Speech Dynamic Database

Acted情感语音动态数据库 (AESDD) 是一个公开可用的语音情感识别数据集。它包含用希腊语表达情感的单词。

创建此数据库的动机是缺乏可用的希腊语SER的高质量数据库，这是在研究戏剧表演的情感触发照明框架时实现的 [1]。数据库表达五种情绪: 愤怒、厌恶、恐惧、快乐和悲伤。

语音情感识别数据集的第一个版本是与一群对拟议框架表现出浓厚兴趣的专业参与者合作创建的。动态 (在AESDD中) 是指通过参与或对项目感兴趣的演员和表演者的贡献来不断扩展数据库的意图。虽然需要做出贡献，但在AESDD上训练的SER模型并不完全以性能为导向。

The Acted Emotional Speech Dynamic Database (AESDD) is a publicly available dataset for speech emotion recognition (SER). It consists of Greek utterances that express emotional states.

The motivation behind developing this database arises from the scarcity of high-quality, publicly available Greek SER datasets. This work was conducted while investigating an emotion-triggered lighting framework for theatrical performances [1]. The database covers five core emotions: anger, disgust, fear, happiness, and sadness.

The first iteration of this SER dataset was developed in collaboration with a cohort of professional participants who demonstrated strong interest in the proposed framework. The term "dynamic" in AESDD refers to the intentional goal of continuously expanding the database through contributions from actors and performers who either participate in the project or express interest in it. While contributions are necessary for the database's ongoing expansion, SER models trained on AESDD are not solely performance-oriented.

搜集汇总

背景与挑战

背景概述

Acted Emotional Speech Dynamic Database (AESDD) 是一个公开的希腊语语音情感识别数据集，包含表达愤怒、厌恶、恐惧、快乐和悲伤五种情绪的语音。该数据集由塞萨洛尼基亚里士多德大学于2018年发布，旨在通过演员的参与动态扩展，用于支持相关研究。