K.R. Ramakrishnan

Publications (3)4.63 Total impact

• Conference Proceeding: A robust initialization scheme for faster convergence of the dichotomous search algorithm for single frequency estimation

  B. Thoshkahna, K.R. Ramakrishnan
  [show abstract] [hide abstract]
  ABSTRACT: The estimation of the frequency of a sinusoidal signal is a well researched problem. In this work we propose an initialization scheme to the popular dichotomous search of the periodogram peak algorithm(DSPA) that is used to estimate the frequency of a sinusoid in white gaussian noise. Our initialization is computationally low cost and gives the same performance as the DSPA, while reducing the number of iterations needed for the fine search stage. We show that our algorithm remains stable as we reduce the number of iterations in the fine search stage. We also compare the performance of our modification to a previous modification of the DSPA and show that we enhance the performance of the algorithm with our initialization technique.
  Signal Processing, 2008. ICSP 2008. 9th International Conference on; 11/2008
• Conference Proceeding: A psychoacoustics based sound onset detection algorithm for polyphonic audio

  B. Thoshkahna, K.R. Ramakrishnan
  [show abstract] [hide abstract]
  ABSTRACT: We propose a new sound onset detection method using principles of psychoacoustics. Our algorithm mimics the human auditory system to detect onsets in polyphonic audio.A filterbank based processing scheme is used that allows for a more robust detection of onsets using a popular model of loudness. We show the versatility of our algorithm by evaluating itpsilas performance over a wide range of sounds.
  Signal Processing, 2008. ICSP 2008. 9th International Conference on; 11/2008
• Conference Proceeding: A Speech-Music Discriminator Using HILN Model Based Features

  B. Thoshkahna, V. Sudha, K.R. Ramakrishnan
  [show abstract] [hide abstract]
  ABSTRACT: We propose a simple speech music discriminator that uses features based on HILN (harmonics, individual lines and noise) model. We have been able to test the strength of the feature set on a standard database of 66 files and get an accuracy of around 97%. We also have tested on sung queries and polyphonic music and have got very good results. The current algorithm is being used to discriminate between sung queries and played (using an instrument like flute) queries for a query by humming (QBH) system currently under development in the lab
  Acoustics, Speech and Signal Processing, 2006. ICASSP 2006 Proceedings. 2006 IEEE International Conference on; 06/2006 · 4.63 Impact Factor