Rosie-Lab/BERSt

BERSt数据集是一个用于各种语音识别任务的语音数据集，包括自动语音识别（ASR）和语音情感识别（SER）。该数据集包含4526条单个短语录音，由98名专业演员在不同家庭环境下使用多种智能手机麦克风录制而成。数据覆盖了不同的英语口音，包括英国、加拿大、美国、澳大利亚等地区以及非英语母语者的口音。数据集中的语音录音涵盖了7种情感类别和3种发声强度级别，还包括覆盖所有英语音素的13条无意义短语。数据集分为训练集、测试集和验证集，且不同集合之间没有说话人交叉。元数据包括演员数量、性别分布、语言使用情况、情感类别计数和距离计数等信息。

The BERSt Dataset is a speech dataset for various speech recognition tasks including Automatic Speech Recognition (ASR) and Speech Emotion Recognition (SER). It contains 4526 single phrase recordings made by 98 professional actors in different home environments using various smartphone microphones. The dataset covers a range of English accents including UK, Canadian, American, Australian, and non-native English accents. The speech recordings span 7 emotion classes and 3 vocal intensity levels, and include 13 nonsense phrases covering all English phonemes. The dataset is split into training, test, and validation sets without speaker crossover between them. Metadata includes details such as the number of actors, gender distribution, language usage, emotion counts, and distance counts.