16th Annual Conference of the International Speech Communication Association

Dresden, Germany
September 6-10, 2015

Measuring and Monitoring Speech Quality for Voice Over IP with POLQA, ViSQOL and P.563

Andrew Hines (1), Eoin Gillen (2), Naomi Harte (2)

(1) Dublin Institute of Technology, Ireland
(2) Trinity College Dublin, Ireland

There are many types of degradation which can occur in Voice over IP (VoIP) calls. Of interest in this work are degradations which occur independently of the codec, hardware or network in use. Specifically, their effect on the subjective and objective quality of the speech is examined. Since no dataset suitable for this purpose exists, a new dataset (TCD-VoIP) has been created and has been made publicly available. The dataset contains speech clips suffering from a range of common call quality degradations, as well as a set of subjective opinion scores on the clips from 24 listeners. The performances of three objective quality metrics: POLQA, ViSQOL and P.563, have been evaluated using the dataset. The results show that full reference metrics are capable of accurately predicting a variety of common VoIP degradations. They also highlight the outstanding need for a wideband, single-ended, no-reference metric to monitor accurately speech quality for degradations common in VoIP scenarios.

Full Paper

Bibliographic reference.  Hines, Andrew / Gillen, Eoin / Harte, Naomi (2015): "Measuring and monitoring speech quality for voice over IP with POLQA, viSQOL and p.563", In INTERSPEECH-2015, 438-442.