/Lecture-Dictation-and-Natural-Language-Querying-of-Knowledge-Corpus

Using Watson STT and Watson Discovery to digitize all the ZOOM/TEAMS recordings from last Academic Year to avoid losing all the work done and turn audio recordings into written, classified and usable persistent knowledge

Primary LanguageJavaScript

Lecture-Dictation-Natural-Language-Querying-of-Knowledge-Corpus

Using Watson STT and Watson Discovery to digitize all the ZOOM/TEAMS recordings from last Academic Year to avoid losing all the work done and turn audio recordings into written, classified and usable persistent knowledge

1.Transcription : Can be done using IBM Watson STT. This should be customized using domains-specific language models which can be done by the students.

2.Ingestion into an NLP-based searchable repository : Using IBM Watson Discovery with trainable queries - the whole set of transcription can be tagged, classified and searchable

#3.Creation of a User Interface that would allow searching of answer passages and links to videos covering particular material : Using the Discovery API a query front-end could be built that would allow - entry of natural language queries, filtering by entities and keywords, links to source video clips.

Deploy to IBM Cloud