Estimation of Glottal Closure Instants in Voiced Speech Using the DYPSA Algorithm

Patrick A. Naylor, Anastasis Kounoudes, Jón Guðnason, Mike Brookes

IEEE Transactions on Audio Speech and Language Processing · 2006 · 315 citations · 34 references

Concepts

TL;DR

Accurate estimation of GCIs is a key tool for speech analysis, synthesis, and coding. The authors introduce DYPSA, an automatic algorithm for estimating glottal closure instants in voiced speech. DYPSA operates solely on the speech signal, using a phase‑slope function and a novel projection technique to generate GCI candidates, then selects the most likely ones via dynamic programming, and is benchmarked against three existing methods. On the APLAWD and SAM databases, DYPSA correctly identifies 95.7 % and 93.1 % of GCIs, respectively.

Abstract

We present the Dynamic Programming Projected Phase-Slope Algorithm (DYPSA) for automatic estimation of glottal closure instants (GCIs) in voiced speech. Accurate estimation of GCIs is an important tool that can be applied to a wide range of speech processing tasks including speech analysis, synthesis and coding. DYPSA is automatic and operates using the speech signal alone without the need for an EGG signal. The algorithm employs the phase-slope function and a novel phase-slope projection technique for estimating GCI candidates from the speech signal. The most likely candidates are then selected using a dynamic programming technique to minimize a cost function that we define. We review and evaluate three existing methods of GCI estimation and compare the new DYPSA algorithm to them. Results are presented for the APLAWD and SAM databases for which 95.7% and 93.1% of GCIs are correctly identified

References

34