Speech Quality of VoIP: Assessment and Prediction
Speech Quality of VoIP: Assessment and Prediction
Multimedia over IP and Wireless Networks: Compression, Networking, and Systems
Multimedia over IP and Wireless Networks: Compression, Networking, and Systems
Objective Assessment of Speech and Audio Quality—Technology and Applications
IEEE Transactions on Audio, Speech, and Language Processing
Voice quality prediction models and their application in VoIP networks
IEEE Transactions on Multimedia
Hi-index | 0.01 |
Available objective Quality of Experience (QoE) assessment methods for speech and audio quality evaluation are not directly usable for quality rating of professional broadband audio communication applications over IP such as audio contribution links for broadcasting. To fill this gap, we designed a dedicated non-intrusive parametric QoE model for conversational quality rating based on the E-model approach. With this the QoE of Audio Contribution over IP (ACIP) can be monitored. Moreover, the estimated QoE scores can be used for a perceptually-driven Quality of Service (QoS) optimization for ACIP, which has different requirements and characteristics compared to Voice over IP (VoIP). In this paper, we present our ACIP QoE model and propose an objective QoE metric for assessing the listening-only quality in ACIP. The latter is used by us for intrusive QoE evaluations, which are necessary for the derivation of a parametric QoE model. Our experimental methodology is illustrated in depth and we give exemplified results. Finally, we demonstrate the application of our model in perceptually-driven QoS optimization.