KIIT College of Engineering conducts world class research in Speech and Language Processing. It has established a research group who is very actively working in several areas of Natural language and speech processing related to Indian spoken languages under the guidance of Dr. S. S. Agrawal, Director General KIIT and an eminent Scientist. The work is going on in collaboration with Indian and Foreign Laboratories. The major areas of research include Automatic Speech Recognition, Speaker identification and development of language speech databases pertaining to Indian languages. Presently, KIIT is conducting advanced research in the following areas:
a) Development of Indian spoken languages Mobile databases
A text corpus of 2 million words of natural messages in 12 different domains in Hindi and Indian English and a spoken corpus of 100 speakers who each spoke 630 phonetically rich sentences, has been created. The speech utterances were recorded in 16 kHz through 3 recording channels: a mobile phone, a headset and a desktop mounted microphone. This project was sponsored by Nokia Research Centre China.
b) Emotional Speech Database
Hindi database for the analysis of isolated words, ten Hindi digits (0-9)were recorded in all six emotions i.e. happy, fear, happy, sad, surprise and neutral, hence a large size corpus of 3000 utterances were created. We are in process for developing the Punjabi, Nepali and Indian English databases for the Department of Ministry and DRDO as per their requirements.
c) Automatic speaker verification and identification using mobile communication data
For this experiment Multilayer error back propagation Feed Forward Neural Network by Associative memory for speaker identification has been considered . In this, 70% of the samples were used to train the network and 15 % were used for validation and 15% for testing of the network. With this specification 20 neurons were used at the hidden layer.
d) Automatic recognition of isolated words
Recognition of Isolated words using Neural Network and Dynamic Time Warping in MATLAB and PRATT tool has been done.
e) Recognition of emotions by Human and Machine
For this experiment six emotions i.e. neutral, happy, sad, fear, anger and surprise has been considered for recognition by human and machine (Neural Network). The acoustic prosodic features such as intensity, duration and intonation changes corresponding to each of these emotions were analyzed using PRATT speech processing software tool. It has been observed that the performance of machine is better as compared with human.
f) Fundamental studies
Acoustic correlation of emotions
Spectrum verification of vowel segments for Indian English, American and Chinese
Comparison of prosodic features
Development of pronunciation lexicon and experimental study of phonetics and phonemics for Punjabi language
Research team is in process of developing trigram language modeling for the purpose of language identification using language specific features
g) Sponsored projects
Projects have been sponsored by MNC like Nokia, China and Government authorities of India like Department of Research and Defense Organization (DRDO) and Department of Ministry (DIT), New Delhi.
h) Foreign Collaboration - Our Research & Development Lab has collaboration with following foreign agencies
i) Research Papers Published (April2011-March2012)
The following papers were published in the national and international conferences related with the work done in KIIT R&D Lab:
S.S. Agrawal ,“Emotions in Hindi Speech-Analysis, Perception and Recognition” O-COCOSDA Oct. 2011, Taiwan.(key note paper)
Shweta Sinha, S.S. Agrawal, Jesper Olsen, “Development of Hindi Mobile communication Text and Speech Corpus” O-COCOSDA Oct.2011, Taiwan.
Archana Balyan, S.S. Agrawal, Amita Dev, “Development of Female Voice Speech Synthesizer for Hindi Language” O-COCOSDA Oct.2011, Taiwan.
Sunita Arora , S.S Agrawal , Karunesh Arora, “Performance of HMM based Hindi Speech Recognizer using Combination of Phoneme and Pause Model” O-COCOSDA Oct. 2011, Taiwan.
S.S.Agrawal, “Comparison of Vowel Spectrum in American, Chinese and Indian English” AESOP workshop Oct. 2011 , Taiwan
Shweta Bansal, Shipra Arora, Manisha Jaitley,Shweta Sinha and S.S. Agrawal , “Acoustic Analysis and Perception of Emotions in Hindi Speech using Words and Sentences” FRSM Jan 2012,KIIT Gurgaon
“Text Independent Speaker Identification for Indian Language by Associative Memory of Close Set” Odyssey March 2012,Singapore. (Submitted)
The Institute is very actively engaged in Research and development in Computer Science and Electronics and Communication areas. It has established a research group who is very actively working in the areas of Natural language and speech processing related to Indian spoken languages under the guidance of Dr. S. S. Agrawal, Executive Director KIIT and an eminent Scientist. Projects related to development of annotated text and speech databases for Hindi and Indian Spoken English from a large population of speakers are being conducted. The work is going on in collaboration with Indian and Foreign Laboratories.
Advance research has been also taken up in Automatic Speech Recognition, Text to Speech Synthesis, Language and Speaker Identification Techniques for Indian Languages. Dr. S. S. Agrawal is also engaged in an international collaborated project called "A- Star" which is coordinated by ATR/NICT, Japan. About 10 different Asian countries including India are participating in this programme.)