vslp2020_vinai_100h_vi_pseudo_labelled
收藏资源简介:
该数据集是一个包含音频及其对应转录文本的数据集,适用于语音识别任务。数据集中的音频采样率为16000Hz,并且每个音频都有其转录文本。此外,还包括了每个音频文件的路径、是否需要依赖前一个样本的条件,以及通过Whisper模型生成的转录文本。整个数据集被划分为训练集、验证集和测试集,分别用于模型的训练、验证和测试。
This dataset contains audio clips and their corresponding transcriptions, tailored for automatic speech recognition (ASR) tasks. All audio files in the dataset have a sampling rate of 16000 Hz, and each audio clip is paired with its transcription. Additionally, the dataset includes the file path of each audio clip, whether the current sample requires contextual information from the prior one, and the transcriptions generated by the Whisper model. The entire dataset is divided into training, validation, and test sets, which are respectively used for model training, validation, and testing.




