A
A. G. Ramakrishnan
Researcher at Indian Institute of Science
Publications - 228
Citations - 3486
A. G. Ramakrishnan is an academic researcher from Indian Institute of Science. The author has contributed to research in topics: Handwriting recognition & Wavelet transform. The author has an hindex of 28, co-authored 220 publications receiving 3111 citations. Previous affiliations of A. G. Ramakrishnan include Birla Institute of Technology and Science.
Papers
More filters
Posted Content
Compressed EEG Acquisition with Limited Channels using Estimated Signal Correlation
TL;DR: It is demonstrated that channels in the 10-10 system of electrode placement can be estimated, with an error less than 10% using recordings on the sparser 10-20 system, and an average error below 15% between the original and reconstructed signals.
Posted Content
Computationally Efficient Approaches for Image Style Transfer.
TL;DR: This work has proposed three modifications to the architecture of a recent, real-time, artistic style transfer technique to make it computationally more efficient, and proposed the use of depth-wise separable convolution (DepSep) in place of convolution and nearest neighbor (NN) interpolation in Place of transposed convolution.
Journal ArticleDOI
Subword Dictionary Learning and Segmentation Techniques for Automatic Speech Recognition in Tamil and Kannada
TL;DR: It is shown that the word error rate (WER) reduces drastically when compared to the baseline word-level ASR, achieving a maximum absolute WER reduction of 6.24% and 6.63% for Tamil and Kannada ASR respectively.
Posted Content
Intrinsic-cum-extrinsic normalization of formant data of vowels
TL;DR: The authors proposed a speaker-extrinsic normalization procedure to re-scales the normalized values by the mean formant values of vowels, which leads to well separated clusters by reducing the spread due to talkers.
Proceedings ArticleDOI
Perceptual-MVDR based analysis-synthesis of pitch synchronous frames for pitch modification
TL;DR: The proposed approach has been rated with higher MOS and has achieved lower MBSD than the time domain pitch synchronous overlap [4], modified-LP method [3] and MVDR based methods.