Signal-based Root Cause Analysis of Quality Impairments in Speech Communication Networks

Konferenz: Speech Communication - 13. ITG-Fachtagung Sprachkommunikation
10.10.2018 - 12.10.2018 in Oldenburg, Deutschland

Tagungsband: ITG-Fb. 282: Speech Communication

Seiten: 5Sprache: EnglischTyp: PDF

Persönliche VDE-Mitglieder erhalten auf diesen Artikel 10% Rabatt

Huebschen, Tobias; Schmidt, Gerhard (Digital Signal Processing and System Theory (DSS), Christian-Albrechts-Universität zu Kiel, Kiel, Germany)
Mittag, Gabriel (Quality and Usability Lab, Technische Universität Berlin, Berlin, Germany)
Moeller, Sebastian (Quality and Usability Lab, Technische Universität Berlin, Berlin, Germany & Language Technology, Deutsches Forschungszentrum für Künstliche Intelligenz (DFKI), Berlin, Germany)

To network providers, identifying the root cause for an observed speech quality degradation in their network is of special interest. In most cases, however, such additional information cannot be provided by state-of-the-art quality assessment algorithms. Therefore, this work proposes an algorithmic system, which is able to identify the individual impact of two commonly simultaneously occurring listening quality degrading root causes, namely speech coding effects and frame errors. For this, the frame error rate, the codec bitrate and the equipment impairment factor of the E-Model are estimated. Results show that these estimates may be reliably used to perform a root cause analysis in network scenarios where the quality degradation caused by the combination of speech coding effects and of frame errors dominates.