Reliable reverberation estimation for improved automatic speech recognition in multi-device systems

摘要

A mechanism for enabling a multi-device reverberation estimation according to an embodiment is described. An apparatus of embodiments as described herein includes detection and recovery logic for allowing a microphone of a first voice-based device of a plurality of voice-based devices to detect a command from a user. The apparatus further includes computational logic for allowing a second voice-based device and a third voice-based device to compute voice-to-reverb modulation power ratio (SRMR) values based on the command, the computational logic further including reverberation times (RTs) based on the To estimate SRMR values. The apparatus further includes a decision and application logic for performing a replay based on the reverberation RT estimated RTs.