Digital Speech Processing Projects

Digital Speech - Digital Speech Processing Projects Algorithmic Projects 1 Speech Detector to separate speech from background sounds noise etc a

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon
Digital Speech Processing Projects Algorithmic Projects: 1. Speech Detector to separate speech from background sounds, noise, etc. a. Based on log energy and zero crossing rate statistics b. Need training set for setting statistical thresholds, testing set to determine algorithm performance 2. Voice/Unvoiced/Silence Detector a. Based on statistical distributions of speech parameters measured from a training set—e.g., published set of 5 parameters b. Need to select appropriate speech features and measure their distributions for all three categories of signal c. Need to measure performance on an independent test set 3. Pitch Detection a. Parallel time domain processor described in class b. Autocorrelation (full band, 900 Hz filtered) c. Average Magnitude Difference Function d. Cepstrum e. Harmonic spectrum i. Need a training set of utterances with known pitch contours to tune algorithm parameters ii. Need a testing set of utterances with known pitch contours to test algorithm performance
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 2
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 12/29/2011 for the course ECE 259 taught by Professor Rabiner,l during the Fall '08 term at UCSB.

Page1 / 2

Digital Speech - Digital Speech Processing Projects Algorithmic Projects 1 Speech Detector to separate speech from background sounds noise etc a

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online