This paper investigates the effects of temporal clipping, which can result from Voice Activity Detection (VAD), or echo canceller's Nonlinear Processor (NLP), on speech quality. Mean Opinion Score (MOS) is used as an index of speech quality. A nonintrusive algorithm, based on the clipping locations, is proposed to predict the speech quality. The algorithm shows excellent performance, the correlation coefficient between the prediction and the measurement is about 0.975, and the root mean square error for the prediction is 0.18 MOS.

Additional Metadata
Keywords Mean opinion score, Nonintrusive method, Speech quality prediction, Temporal clipping, Voice activity detection, Voice over internet protocol
Persistent URL dx.doi.org/10.1109/IMTC.2004.1351264
Conference IMTC'05 - Proceedings of the IEEE Instrumentation and Measurement Technology Conference
Citation
Lijing, D. (Ding), El-Hennawey, M.S. (Mohamed Samy), & Goubran, R. (2005). Measurement of the effects of temporal clipping on speech quality. In Conference Record - IEEE Instrumentation and Measurement Technology Conference (pp. 1135–1138). doi:10.1109/IMTC.2004.1351264