This dataset consists of agriculture videos in hindi and oriya.
The dataset consists of mulitple xls files and each xls file has column of video urls (youtube video links) and corresponding transcripts.
The transcripts are:
Generated by ASR models (for the purpose of benchmarking)
Manual transcripts
Time stamps
Manual translations