kotoba-speech/whisper_transcriptions.reazonspeech.all.wer_10.0.vectorized.dictionary

该数据集包含了多种配置的训练数据，每种配置下都有input_length、labels、input_features、labels.text和keywords等字段。数据集适用于训练机器学习模型，其中input_length表示输入序列的长度，labels表示标签，input_features表示输入特征，labels.text表示标签文本，keywords表示关键词。每种配置的训练集大小、示例数量、下载大小和总数据大小不同。

The dataset contains training data for multiple configurations, each with fields such as input_length, labels, input_features, labels.text, and keywords. The dataset is suitable for training machine learning models, where input_length represents the length of the input sequence, labels represent the tags, input_features represent the input features, labels.text represents the tag text, and keywords represent the keywords. The training set size, number of examples, download size, and total data size differ for each configuration.