The ODSQA dataset is a spoken dataset for question answering in Chinese. It contains more than three thousand questions from 20 different speakers.
3 PAPERS • NO BENCHMARKS YET