IISc-MILE Tamil ASR Corpus是一个用于自动语音识别(ASR)的泰米尔语转录语音语料库。该数据集由专家生成,语言为泰米尔语,许可证为CC BY 2.0,属于单语言数据集,大小在10K到100K之间。数据集的源数据为原始数据,标签包括Tamil ASR和Speech Recognition,任务类别为automatic-speech-recognition。
Introduction The CALLFRIEND project supports the development of language identification technology. Data The corpus consists of 60 unscripted telephone conversations, lasti