SpeechPPL/SALMon_GSLM-greedy

这是一个包含音频数据和相关信息的的数据集，用于某种任务类型的训练。数据集包括正例音频、负例音频、过渡音频、提示音频、正例延续音频、负例延续音频、负例音频稳健性检查、各类样本的逐标记损失、模型生成的延续音频、代码帧率、代码深度和模型采样率等特征。数据集分为训练集，共有200个示例。

This dataset contains audio data and related information for training on a certain task type. It includes features such as positive audio, negative audio, transition audio, prompt audio, positive continuation audio, negative continuation audio, negative audio sanity check, token-wise loss for various samples, model-generated continuation audio, code frame rate, code depth, and model sampling rate. The dataset is split into a training set with a total of 200 examples.