Automatic Speech Recognition Using Context-Dependent Modelling of Deep Neural Networks Using Logistic Regression (13321N)
IDA Technology Roadmap 2012
This technology falls in the following categories of Singapore's IDA Infocomm Technology Roadmap 2012:
For more information on this technology contact:
The invention is based on a regression model that allows finer modelling of the acoustic contexts without dramatically increasing the model complexity. This technology is built on top of the state-of-the-art deep learning paradigm. By defining the canonical acoustic contexts based on the articulator attributes, the algorithm learns a regression function that can better recognise continuous speech sounds. This learning technique is especially useful in conditions where there is insufficient training data to cover different acoustic contexts.
Technology Readiness Level 4 on the scale by the Ministry of Defence Singapore.
About the Research Group
Dr. Khe Chai Sim is an Assistant Professor at the National University of Singapore (NUS). His research focuses on automatic speech recognition, acoustic keyword spotting and other areas of speech processing.