Related Experiment Video
Updated: Apr 13, 2026
Foreign Accent and Forensic Speaker Identification in Voice Lineups: The Influence of Acoustic Features Based on Prosody
Published on: September 27, 2024
Comparability of Computer-Supported Concurrent Voice Analysis
Holger Hanschmann1, Sebastian Gärtner, Roswitha Berger
1Private Practice for ENT, Allergology, Phoniatrics and Pedaudiology, Frankfurt am Main, Germany.
Objective:
The assessment of vocal ability parameters can be made easier and more objective by computer-supported analysis systems. For this, multiple commercially available systems exist. But how strongly do the measured results of one method differ from the results of another and is the evaluation result of one method replaceable by or directly comparable with another?
Patients And Methods:
15 women aged 20-55 years and 15 men aged 18-56 years took part in this prospective study. The programs lingWAVES and rpSzene® were compared for the results of jitter, shimmer and glottal-to-noise excitation ratio. Correlation coefficients, Youden plots and Bland-Altman plots were calculated to make the comparison.
Results:
For both systems, none of the values were in accordance with a normal distribution. The results of the two programs did not produce satisfactory consistency.
Conclusions:
In this study, absolute equivalency in the recording was ensured by the simultaneous recording of the voice signal and subsequent analysis with two different voice analysis methods. Nevertheless, the results varied so much that they cannot be compared properly. To reach a comparability of voice analyses, it is necessary to develop perturbation measurements which guarantee good clinical applicability and comparability between different analysis systems.
