• Sentence Parsing and extraction of spatial relation for human robot interaction:

Techniques Used: POS tagging, Named Entity Extraction, Anaphora Resolution, Stanford Parser, Spacy dependency parser .

Platform: Python, NLTK, Stanford NER.

• Negative news scanning (Sentiment Analysis):

Techniques Used: POS tagging, Named Entity Extraction, Anaphora Resolution, Bayes (Multinomial, unigram).

Platform: Python, NLTK, Stanford NER.

• Chatbot design:

Techniques Used: POS tagging, Bayes (Multinomial, unigram).

Platform: Python, NLTK, Chatterbot, JSON.

• Grammer generation and acoustic modelling:

Techniques Used: finite state transducer, automata theory

Platform: Python, Amazon web service, shell scripting, putty, pyfst, openfst.

• Gesture Recognition:

Techniques Used: Wavelet descriptor, orientation histogram, hidden markov model, possibility theory, HSV colour segmentation.

Platform: MATLAB, Open CV, Python

• Speech Recognition:

Techniques Used: Discrete Wavelet Transform, MFCC, HFCC, hidden markov model, possibility theory.

Platform: MATLAB, Voice Box

Isolated Indian Sign Language Dataset

Continuous Indian Sign Language Dataset

If you use our dataset for publishing results, please cite the following paper:

1- Neha Baranwal and G.C.Nandi.” An Efficient Gesture based Humanoid Learning using Wavelet Descriptor and MFCC Techniques”. Published in International Journal of Machine Learning and Cybernetics (Springer) (SCI) (2.692 impact factor) (Vol. 8, issue 4, pp. 1369-1388, 17-april-2016)( DOI:10.1007/s13042-016-0512-4).

2- Neha Baranwal and G.C.Nandi "Development of a Framework for Human-Robot interactions with Indian Sign Language Using Possibility theory" published International journal of social robotics (springer) (SCI) (2.009 impact factor)(30-05-2017)(DOI 10.1007/s12369-017-0412-0).