What is it about?
Many people with voice problems describe their voice as hoarse, rough, breathy, or weak. Clinicians often try to measure these changes using computer analysis of recorded speech. But some common voice measurements can become unreliable when the voice is very irregular. This article explains a more reliable approach called cepstral analysis, in simple terms, and focuses on two widely used measures (often shortened to CPP and CPPS). We describe what these measures are trying to capture, essentially, how “clear and steady” the voice signal is compared with background noise or irregular vibration, and how the results can change depending on what a person says (holding a vowel like “aaa” versus speaking in sentences) and how the measurement is computed in different software. We also summarize research findings on how these measures relate to voice disorder severity and tracking improvement with treatment, and we compile typical reference values when studies report them.
Featured Image
Photo by Vika Strawberrika on Unsplash
Why is it important?
What makes this paper unique is that it turns cepstral measures (CPP/CPPS) from “numbers on a screen” into something clinicians, students, and researchers can use confidently. It explains the concepts in more comprehensible terms and, importantly, clarifies why CPP/CPPS results can change across software, versions, and settings, sometimes by a large amount, so people don’t apply the wrong cutoff or compare results unfairly. This is timely because cepstral analysis is now widely used (including in more automated and remote workflows), and the paper can improve consistency, reduce misinterpretation, and support more reliable voice assessment and treatment tracking.
Perspectives
As a student and a voice clinician, I genuinely struggled to understand cepstral analysis because it often depends on signal-processing concepts, calculations, and algorithms that are not explained in a clinician-friendly way in the literature. While a few pioneering papers introduced CPP/CPPS, the explanations were still largely technical. I’m grateful to my PhD supervisor for taking the time to teach me these concepts with precision and clarity, and this review is my way of sharing what I learned so others can access it more easily. Our key message is that CPP/CPPS are powerful, clinically validated measures, but only when interpreted in the right context (task choice, software/version, and parameter settings). A critical takeaway is that absolute values can differ substantially across platforms and versions (sometimes by ~7–10 dB), so clinicians should avoid copy-pasting cutoffs from unrelated tasks or software. Ultimately, I hope this paper helps clinicians and students use cepstral measures more confidently by combining clear explanations with practical guidance on how to extract, report, and interpret CPP/CPPS responsibly.
Ketaki Vasant Phadke
Read the Original
This page is a summary of: A Narrative Review on Cepstral Analysis of Voice: Simplified Concepts and Applications, Perspectives of the ASHA Special Interest Groups, December 2025, American Speech-Language-Hearing Association (ASHA),
DOI: 10.1044/2025_persp-25-00091.
You can read the full text:
Contributors
The following have contributed to this page
