On the quality-assessment of reverberated speech

This paper addresses the problem of quantifying the reverberation effect in speech signals. The perception of reverberation is assessed based on a new measure combining the characteristics of reverberation time, room spectral variance, and direct-to-reverberant energy ratio, which are estimated from the associated room impulse response (RIR). The practical aspects behind a robust RIR estimation are underlined, allowing an effective feature extraction for reverberation evaluation. The resulting objective metric achieves a correlation factor of about 90% with the subjective scores of two distinct speech databases, illustrating the system’s ability to assess the reverberation effect in a reliable manner.

► We addressed the problem of quantifying the reverberation effect in speech signals. ► The proposed measure combines three characteristics of the room, T60, RSV and Ed/Er. ► The quality assessment system consists in estimating RIR and its characteristics. ► It was tested using two databases, reaching 91% and 95% of correlation with MOS. ► The results outperformed other reverberation-assessing methods in the literature.