Online Blind Reverberation Time Estimation Using CRNNs

The reverberation time, T-60, is an important acoustic parameter in speech and acoustic signal processing. Often, the T-60 is unknown and blind estimation from a single-channel measurement is required. State-of-the-art T-60 estimation is achieved by a convolutional neural network (CNN) which maps a feature representation of the speech to the T-60. The temporal input length of the CNN is fixed. Time-varying scenarios, e.g., robot audition, require continuous T-60 estimation in an online fashion, which is computationally heavy using the CNN. We propose to use a convolutional recurrent neural network (CRNN) for blind T-60 estimation as it combines the parametric efficiency of CNNs with the online estimation of recurrent neural networks and, in contrast to CNNs, can process time-sequences of variable length. We evaluated the proposed CRNN on the Acoustic Characterization of Environments Challenge dataset for different input lengths. Our proposed method outperforms the state-of-the-art CNN approach even for shorter inputs at the cost of more trainable parameters.