Reverberation is a common problem for many speech technologies, such as automatic speech recognition (ASR) systems. This paper investigates the novel combination of precedence, binaural and statistical independence cues for enhancing reverberant speech, prior to ASR, under these adverse acoustical conditions when two microphone signals are available. Results of the enhancement are evaluated in terms of relevant signal measures and accuracy for both English and Polish ASR tasks. These show inconsistencies between the signal and recognition measures, although in recognition the proposed method consistently outperforms all other combinations and the spectral-subtraction baseline.

The presented paper concerns the issues of communication networks applied to monitoring and control of reactive power compensator for small hydroelectric plants installed in areas distant from urban agglomerations. Ethernet, CAN, Modbus and GPRS transmission protocols has been used. Industrial programmable controller as a data collector has been used also.