Header menu link for other important links
Implementation of pitch detection algorithms for pathological voices
K. Kolhatkar, M. Kolte,
Published in Institute of Electrical and Electronics Engineers Inc.
Volume: 1
Pitch detection algorithm is important feature in the speech processing algorithms. It can be used for speaker recognition, speech instruction of the hearing impaired ones, vocoder systems, discriminate normal and pathological voice etc. Hence robust and accurate determination of pitch is necessary. The paper involves determining pitch of the speech signal by the pitch detection algorithm - 1) Autocorrelation method, 2) Cepstrum Method, 3) Simplified Inverse Filtering Method 4) Data Reduction Method. A measurement is made on the pitch contours to explore how voice/acoustic parameter like fundamental frequency can determine person's health. The paper attempts to identify and classify the pathological voices from the normal voices using k-NN algorithm. The relative performance comparison is done in terms of accuracy and speed of operation of the above mentioned algorithms.