Some of the projects listed below are college projects and some are on my own time.
This project is about generating the news headline after giving seeding two or three words. The dataset is scraped from the Nepal most popular news TV channel website Ekantipur which contains the headline of the news. And the word dictionary is build on top of the words present in those headlines. The Model is deployed on the Streamlit for demo purposes.
Demo LinkTags:
NLPDeep LearningTensorflowText GenerationStreamlitA model which accepts recorded audio as input and output the words spoken in the audio along with the metrics such as Word Error Rate (WER) and Character Error Rate(CER). We used transfer learning on the pre-trained model provided by the Google - Transformers on the open source LibriSpeech Audio Dataset which has approximately of 1000 hours and is sampled at the rate of 16KHz.
Demo LinkTags:
Deep LearningTensorflowSpeech RecognitionLibriSpeech DatasetThis was our minor project in the college senior years in which we used the most popular MERN stack to develop a web app which can help students ask question related to their curriculum and get answers from the peers, seniors or teachers. It also had a fund raising feature as we integrated the Nepal digital wallet - Khalti so that visitors who get their question answered can donate some amount like 10 or so.
Demo LinkTags:
Mongo DBExpressReactNode.jsThis was my own software engineering project in which i develop a web app that allowed students of my campus - Western Region Campus to sell and buy used books and notes. Furthermore, it also provided free pdfs of all faculties for the students to download. .
Demo LinkTags:
ExpressHandlebarsMongoDBNode.js