Short- and Long-Term Packet Loss Behavior: Towards Speech Quality Prediction for Arbitrary Loss Distributions

A speech-quality-oriented classification of packet loss distributions is proposed according to both the short- and long-term loss behavior. While the short-term behavior (microscopic loss behavior) relates to the effect of packet loss on the coder and packet loss concealment performance, the long-term loss behavior (macroscopic loss behavior) is defined so that it reflects the loss behavior that ultimately leads to speech quality that perceptively changes over time. Based on this classification, different parametric (objective) modeling approaches for predicting speech quality are discussed. To this aim, a packet loss averaging approach is presented for modeling speech quality under short-term loss. Starting from this model, two different ways for predicting speech quality under long-term-dependent packet loss are analyzed and compared to auditory (subjective) test results: quality prediction based on the averaging at packet trace level as provided, for example, by the E-model (2005), and the prediction based on the time-averaging of estimated instantaneous quality profiles, as suggested, for example, by L. Gros and N. Chateau (2001) (1998). From this comparison, the suitability of the different approaches for network planning are discussed, and their limitations in case of particular loss distributions are pointed out

[1]  Vincent Barriac,et al.  ARCHITECTURE OF NON-INTRUSIVE PERCEIVED VOICE QUALITY ASSESSMENT , 2001 .

[2]  Sebastian Möller,et al.  Impairment Factor Framework for Wide-Band Speech Codecs , 2006, IEEE Transactions on Audio, Speech, and Language Processing.

[3]  Alexander Raake,et al.  Elements of interactivity in telephone conversations , 2004, INTERSPEECH.

[4]  Scott Pennock.  Accuracy of the Perceptual Evaluation of Speech Quality (PESQ) algorithm , 2002 .

[5]  Alexander Raake.  Predicting speech quality under random packet loss: Individual impairment and additivity with other network impairments , 2004 .

[6]  Phil Gray,et al.  An Experimental Investigation of the Accumulation of Perceived Error in Time-Varying Speech Distortions , 1997 .

[7]  M. Hansen,et al.  Continuous assessment of time-varying speech quality. , 1999, The Journal of the Acoustical Society of America.

[8]  Jan Skoglund,et al.  iLBC - a linear predictive coder with robustness to packet losses , 2002, Speech Coding, 2002, IEEE Workshop Proceedings..

[9]  METHODS FOR SUBJECTIVE DETERMINATION OF TRANSMISSION QUALITY Summary , 2022 .

[10]  Wenyu Jiang,et al.  Comparison and optimization of packet loss repair methods on VoIP perceived quality under bursty loss , 2002, NOSSDAV '02.

[11]  Georg Carle,et al.  Framework model for packet loss metrics based on loss runlengths , 1999, Electronic Imaging.

[12]  Sebastian Möller,et al.  Assessment and Prediction of Speech Quality in Telecommunications , 2000 .

[13]  Alexander Raake,et al.  Speech Quality of VoIP - Assessment and Prediction , 2006 .

[14]  Alexander Raake,et al.  The well-tempered conversation: interactivity, delay and perceptual VoIP quality , 2005, IEEE International Conference on Communications, 2005. ICC 2005. 2005.

[15]  Guido H. Petit,et al.  Assessing Voice Quality in Packet-Based Telephony , 2002, IEEE Internet Comput..

[16]  Donald F. Towsley,et al.  Adaptive FEC-based error control for Internet telephony , 1999, IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320).

[17]  Robert G. Cole,et al.  Voice over IP performance monitoring , 2001, CCRV.

[18]  Alan Clark,et al.  Modeling the effects of burst packet loss and recency on subjective voice quality , 2001 .

[19]  Ville-Veikko Mattila.  Ideal Point Modelling of Speech Quality in Mobile Communications Based on Multidimensional Scaling (MDS) , 2002 .