The well-tempered conversation: interactivity, delay and perceptual VoIP quality

The factors causing perceptual quality impairment on voice-over-IP (VoIP) connections include traditional network quality-of-service (QoS) parameters like packet loss rate, delay or jitter as well as parameters characterizing the conversation itself. Among the latter ones, we focus on the impact of "conversational interactivity" on the perceptual quality of a phone conversation. We introduce "parametric conversation analysis" as a formal framework for the instrumental investigation of conversational parameters at different transmission delay conditions, we further present the notion of "conversational temperature" as an intuitive scalar metric for the interactivity of conversations, and we demonstrate the application of our methods to a set of conversation recordings performed under various delay conditions, also with respect to results of subjective quality ratings.

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

[2]  Nobuhiko Kitawaki,et al.  Pure Delay Effects on Speech Quality in Telecommunications , 1991, IEEE J. Sel. Areas Commun..

[3]  D. Evans.  Introduction to statistical mechanics and thermodynamics , 1984 .

[4]  Sheldon M. Ross,et al.  Stochastic Processes , 2018, Gauge Integral Structures for Stochastic Calculus and Quantum Electrodynamics.

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

[6]  E. Schegloff,et al.  A simplest systematics for the organization of turn-taking for conversation , 1974 .

[7]  E. Heath.  Borg's Perceived Exertion and Pain Scales , 1998 .

[8]  Peter Reichl,et al.  Hot discussion or frosty dialogue? towards a temperature metric for conversational interactivity , 2004, INTERSPEECH.

[9]  M. Tosi.  Introduction to statistical mechanics and thermodynamics , 1997 .

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

[11]  Paul T. Brady,et al.  A statistical analysis of on-off patterns in 16 conversations , 1968 .

[12]  Akira Takahashi,et al.  Opinion model for estimating conversational quality of VoIP , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[13]  Karen Ruhleder,et al.  Meaning-making across remote sites: How delays in transmission affect interaction , 1999, ECSCW.

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