Accession Number:

ADA127671

Title:

A Post-Processing Algorithm for Time Domain Pitch Trackers.

Descriptive Note:

Interim rept.,

Corporate Author:

CARNEGIE-MELLON UNIV PITTSBURGH PA DEPT OF COMPUTER SCIENCE

Personal Author(s):

Report Date:

1983-01-21

Pagination or Media Count:

22.0

Abstract:

This paper describes a powerful post-processing algorithm for time-domain pitch trackers. On two successive passes, the post-processing algorithm eliminates errors produced during a first pass by a time-domain pitch tracker. During the second pass, incorrect pitch values are detected as outliers by computing the distribution of values over a sliding 80 msec window. During the third pass based on artificial intelligence techniques, remaining pitch pulses are used as anchor points to reconstruct the pitch train from the original waveform. The algorithm produced a decrease in the error rate from 21 obtained with the original time domain pitch tracker to 2 for isolated words and sentences produced in an office environment by 3 male and 3 female talkers. In a noisy computer room errors decreased from 52 to 2.9 for the same stimuli produced by 2 male talkers. The algorithm is efficient, accurate, and resistant to noise. The fundamental frequency micro-structure is tracked sufficiently well to be used in extracting phonetic features in a feature-based recognition system. Author

Subject Categories:

  • Non-Radio Communications

Distribution Statement:

APPROVED FOR PUBLIC RELEASE