Training/Evaluation scripts for pipeline baseline in SLUE QA benchmark proposed in the paper SLUE Phase-2: A Benchmark Suite of Diverse Spoken Language Understanding Tasks
*** Note: these instructions are a work in progress and will be updated over the next few days
Fine-tune the text QA model with the groundtruth transcripts of the SLUE-SQA-5 fine-tune set
sh train.sh
Evaluate the text QA model with the groundtruth transcripts of the SLUE-SQA-5 dev set
sh eval_groundtruth.sh