Our research focus is to address issues in processing, analyzing, understanding and manipulation of “real” speech i.e., read, conversational, emotional as found in real-environments for the development of robust speech systems in multiple languages specifically in the context of India. The speech group is also focused on development of speech recognition, speech synthesis, prosody models, speech enhancement, spoken audio search, phonetic engine for Indian languages, language identification, speaker recognition for biometrics, voice conversion, speech summarization and spoken dialog systems.
The lab is headed by Dr. Sri Rama Murty Kodukula.
Acoustic modelling for speech to text
Analysis of acoustic environments for audition research
English to Indian language translations
Multi microphone array based enhancement strategies
Generation of speech waveform closest to human