End-to-End Indian English Automatic Speech Recognition (ASR) systems have been developed for different domains like news, stories, articles and NPTEL lecture transcription domains like Humanities, Electrical Engineering, Electrical and Communication Engineering, Computer Science Engineering and Mechanical Engineering. A Speech activity detector is developed for distinguishing speech and silence. The segment of speech thus detected is used by the ASR systems to generate the transcriptions automatically. This automatically generated transcription is used in generating Subtitles / Notes.  

ASR has been provided as an offline service. Users have to upload wav files (not more than 200MB) at this link: https://www.iitm.ac.in/speech/NPTEL/audio/ The link to download the ASR output will be shared in the same website. The average turnaround time is 24 hours.