Abstract: Embodiments of a method and apparatus for processing a speech are provided. The method can include: acquiring, in response to determining at least one speech interaction device in a target speech interaction device set receiving an input speech, a speech feature of the input speech received by a speech interaction device of the at least one speech interaction device; and selecting, based on the speech feature of the input speech received by the speech interaction device in the at least one speech interaction device, a first speech interaction device from the at least one speech interaction device to process the input speech. Some embodiments realize the selection of a targeted speech interaction device.

Abstract: Embodiments of a method and apparatus for processing a speech are provided. The method can include: acquiring, in response to determining at least one speech interaction device in a target speech interaction device set receiving an input speech, a speech feature of the input speech received by a speech interaction device of the at least one speech interaction device; and selecting, based on the speech feature of the input speech received by the speech interaction device in the at least one speech interaction device, a first speech interaction device from the at least one speech interaction device to process the input speech. Some embodiments realize the selection of a targeted speech interaction device.

Abstract: A tool condition monitoring system monitors the power consumption of a tool during performance of a cyclical task. The system monitors a plurality of characteristics of the power consumption and diagnoses the condition of the tool based upon the plurality of characteristics. A rule base of the plurality of characteristics is generated in a learning mode by monitoring a tool of known condition for a plurality of cycles. In monitor mode, the system monitors the power consumption of a tool and diagnosis the condition of the tool by indexing the rule base based upon the plurality of characteristics monitored.

Abstract: A tool condition monitoring system monitors the power consumption of a tool during performance of a cyclical task. The system monitors a plurality of characteristics of the power consumption and diagnoses the condition of the tool based upon the plurality of characteristics. A rule base of the plurality of characteristics is generated in a learning mode by acquiring a tool of known condition for a plurality of cycles. In monitor mode, the system monitors the power consumption of a tool and diagnosis the condition of the tool by indexing the rule base based upon the plurality of characteristics monitored. In addition, a method is disclosed wherein the band may be modified for accommodating tool wear and avoiding nuisance detection such that the system can keep monitoring in tool's lifetime.

Abstract: A tool monitoring system monitors the condition of an electrically powered tool performing a cyclical operation. The tool monitoring system operates generally in two modes: learning mode and monitor mode. In learning mode, the tool monitoring system gathers statistical data on the power consumption of tools of the selected tool type during learning cycles. A power threshold is generated based upon the statistical data. The tool monitoring system then counts the number of crossings by each of the learning cycles of the power threshold and generates statistical data regarding the number of crossings. Preferably, the mathematical operation of wavelet packet transform is used to calculate the power threshold. Feature wavelet packets of the power consumption signal of the tool are calculated. The power consumption signal is then reconstructed from the feature wavelet packets and used to determine the power threshold.