Abstract: A method of building a new voice having a new timbre using a timbre vector space includes receiving timbre data filtered using a temporal receptive field. The timbre data is mapped in the timbre vector space. The timbre data is related to a plurality of different voices. Each of the plurality of different voices has respective timbre data in the timbre vector space. The method builds the new timbre using the timbre data of the plurality of different voices using a machine learning system.

Abstract: A method for online voice content moderation provides a multi-stage voice content analysis system. The system includes a pre-moderator stage having a toxicity scorer configured to provide a toxicity score for a given toxic speech content from a user. The toxicity score is a function of a platform content policy. The method generates a toxicity score for the given toxic speech content. The toxic speech content is provided to a moderator as a function of the toxicity score.

Abstract: A fly-back or boost stage transfers its stored energy to the energy storage capacitor of the fast driver discharge stage in a single event pulse. The charging voltage of a single flyback pulse on the capacitor is measured in real time and, if necessary, the charging current is diverted via a shunt active device or transistor, to ground, thus preventing the storage capacitor from overcharging with the risk of component damage. A series sense resistor is used to determine the presence and amount of the wasted shunt current in order that this current may be minimized by turning down the flyback energy, thus maximizing the overall efficiency.

Abstract: A method watermarks speech data by using a generator to generate speech data including a watermark. The generator is trained to generate the speech data including the watermark. The training process generates first speech from the generator. The first speech data is configured to represent speech. The first speech data includes a candidate watermark. The training also produces an inconsistency message as a function of at least one difference between the first speech data and at least authentic speech data. The training further includes transforming the first speech data, including the candidate watermark, using a watermark robustness module to produce transformed speech data including a transformed candidate watermark. The transformed speech data includes a transformed candidate watermark. The training further produces a watermark-detectability message, using a watermark detection machine learning system, relating to one or more desirable watermark features of the transformed candidate watermark.

Abstract: A cable management system for an overhead EVSE employs a lever-latch on the EV connector. A controller and motor mechanism is employed to lower the connector to an ADA compliant position or a pre-established height which is typically four feet. For an overhead system which employs a shuttle which slides along a track, the connector may then be grasped to pull the shuttle to a selected position along the track. The latch is depressible to extend the service cable to the connector so that the connector may be positioned and connected at the EV inlet. When the connector is unlatched from the EV inlet, the latch may be depressed to retract the connector to the pre-established height or ADA position. The latch requires a force of less than 5 lbs. for activation.

Abstract: A method of building a new voice having a new timbre using a timbre vector space includes receiving timbre data filtered using a temporal receptive field. The timbre data is mapped in the timbre vector space. The timbre data is related to a plurality of different voices. Each of the plurality of different voices has respective timbre data in the timbre vector space. The method builds the new timbre using the timbre data of the plurality of different voices using a machine learning system.

Abstract: A timbre vector space construction system for building a timbre vector space has an input. The input is configured to receive a first speech segment in a first voice and a second speech segment in a second voice. The system also includes a temporal receptive field to transform the first speech segment into a first plurality of analytical segments, and the second speech segment into a second plurality of analytical segments. Each of the first plurality of smaller analytical segments, and each of the second plurality of analytical segments have a frequency distribution that represents a different portion of the timbre data of the respective voices. The system also includes a machine learning system configured to map the first voice relative to the second voice in the timbre vector space as a function of the frequency distribution of the first plurality of analytical segments the second plurality of analytical segments.

Abstract: A current driver is disclosed which allows very short pulses at high currents to be generated for high power laser diodes. The parasitic inductance of the laser diode limits the speed at which the laser diode may be turned on and off. A high voltage is used to charge this inductance rapidly and maximize the rise time. The fall time is shortened by allowing a similar high voltage to be generated at turnoff without damage to the laser diode or switching components. A portion of the energy stored in the parasitic inductance may recovered to reduce drain on the power source, and to improve overall efficiency. The anode of the laser may be switched to ground at the end of a pulse.

Abstract: A method of building a new voice having a new timbre using a timbre vector space includes receiving timbre data filtered using a temporal receptive field. The timbre data is mapped in the timbre vector space. The timbre data is related to a plurality of different voices. Each of the plurality of different voices has respective timbre data in the timbre vector space. The method builds the new timbre using the timbre data of the plurality of different voices using a machine learning system.

Abstract: A method of building a speech conversion system uses target information from a target voice and source speech data. The method receives the source speech data and the target timbre data, which is within a timbre space. A generator produces first candidate data as a function of the source speech data and the target timbre data. A discriminator compares the first candidate data to the target timbre data with reference to timbre data of a plurality of different voices. The discriminator determines inconsistencies between the first candidate data and the target timbre data. The discriminator produces an inconsistency message containing information relating to the inconsistencies. The inconsistency message is fed back to the generator, and the generator produces a second candidate data. The target timbre data in the timbre space is refined using information produced by the generator and/or discriminator as a result of the feeding back.

Abstract: A current driver is disclosed which allows very short pulses at high currents to be generated for high power laser diodes. The parasitic inductance of the laser diode limits the speed at which the laser diode may be turned on and off. A high voltage is used to charge this inductance rapidly and maximize the rise time. The fall time is shortened by allowing a similar high voltage to be generated at turnoff without damage to the laser diode or switching components. A portion of the energy stored in the parasitic inductance may recovered to reduce drain on the power source, and to improve overall efficiency. The anode of the laser may be switched to ground at the end of a pulse.

Abstract: A multiple EVSE installation employs a power sharing system for EVSE pairs which share a common circuit breaker. When only one EVSE is connected to an electric vehicle for charging, the connected EVSE is allowed to charge at full capacity. When a second EV connects to the second EVSE of the pair, while the first EVSE is charging, a controller senses the latter event and issues commands to both EVSE units to limit their current so that the rated circuit breaker capacity is not exceeded. The controller continuously monitors the charging of each of the EVSE units and sends commands to adjust the supply of current to each EVSE accordingly.

Abstract: A technique for is presented for locating at least one object illuminated by a laser designator. A detector and an optical unit are provided. The one optical unit is configured for receiving a beam of laser light scattered by the at least one object being illuminated by the laser designator, for creating two secondary beams, and for focusing the two secondary beams to respective foci along an optical axis of the at least one optical unit. The detector is located between the two foci, divided into an even number of portions, and is configured for measuring an intensity of impinging light at each portion. The optical unit is configured for causing the received laser light to impinge the detector, such that an upright image and an inverted image of the scattered laser light beam having substantially equal sizes and substantially equal energies are projected on the detector.

Abstract: A system involves the use of one or more sensor pins which generate a binary output to determine the position of the tire on the sensor module, and if the depth of the tire tread meets the recommended depth for safe operation. The vehicle drives over two linear sensor modules, one for the driver side tires and the other for the passenger side tires. As the tires pass over the sensor modules, the sensor pins on/off status is recorded and analyzed to determine whether the tires are centered on the module and whether the depth of the threads meet or exceed the recommended depth for safe operation. When both front and rear tires have passed over the sensor modules, the tire statuses will be displayed on a tire status indicator, and a receipt with the tire status will be printed for the vehicle. When used with fleet management systems, the vehicle identification and the tire statuses will also be sent to a data center for further processing.

Abstract: An overhead cable management for charging an electric vehicle employs a reel which receives a cable with an EV connector. A drive assembly employs a bi-positionable clutch and a gear assembly which sequentially engages a drive gear for the reel to unwind the cable from an overhead position to an ADA height, allow the cable and connector to be extended to connect with the EV and to wind the cable until a locked home position is obtained. The cable management is incorporated into an EVSE which is ceiling mounted, or an EVSE which is mounted to a pole or a wall. Numerous modules and features are optionally employed in connection with the operation of the cable management.

Abstract: An electric charging system for charging electric vehicles is particularly adapted for use in a parking garage. Parking spaces in the garage are fixed with a charging station so that the vehicles may be charged while they park at the garage. The power cable which connects to charge the vehicle battery is suspended at a height generally above the vehicle and is automatically retractable. Information concerning the power charge is automatically transmitted to the payment station at the exit gate. A point of sale (POS) module is mounted for communication with the power cable. The point of sale (POS) module and the connector are easily accessible and the connector is positionable for ease of connection with the vehicle terminal. A retractor for extending and retracting an overhead power cable employs a drive wheel and a clutch which are electronically controlled.

Abstract: A connection line protection system employs unique identifiers on outlets and inlets. A connector connects a selected outlet with a selected inlet. The connector mounts at opposing ends an identifier that identifies the connector. A scanner is employed to scan in tandem a first identifier of the outlet and a connector identifier and a second identifier of the inlets and the identifier of the connector. The readings are compared with a schedule to determine whether the connector is properly mounted and connected between a selected outlet and a selected inlet. The identifiers include RFIDs and barcodes. The identifiers are preferably angularly spaced around the outlets, inlets and the couplers of the connector. The identifier may be affixed by snap rings mounting barcode labels or RFID buttons.

Abstract: A technique is presented for tracking a target illuminated by a laser designator. A light beam scattered from the target in response to illumination by the designator is received and projected on a detector as an unfocused spot. The detector is divided into a plurality of sectors, each sector being adjacent to two sectors and all sectors meeting at a common point. Each sector outputs a signal indicative of a respective energy of the portion of the light beam illuminating the sector of the detector. A sum of the signals is determined. A plurality of ratios is determined, each ratio corresponding to a respective sector and being calculated by dividing a signal from the respective sector by the sum of the signals. The orientation of the system with respect to the target is determined based on the three ratios, via a look-up table and/or via one or more algorithms.

Abstract: A multiple EVSE installation employs a power sharing system for EVSE pairs which share a common circuit breaker. When only one EVSE is connected to an electric vehicle for charging, the connected EVSE is allowed to charge at full capacity. When a second EV connects to the second EVSE of the pair, while the first EVSE is charging, a controller senses the latter event and issues commands to both EVSE units to limit their current so that the rated circuit breaker capacity is not exceeded. The controller continuously monitors the charging of each of the EVSE units and sends commands to adjust the supply of current to each EVSE accordingly.

Abstract: A fuel dispensing installation employs multiple fuel islands with fuel dispensers and multiple fuel/travel lanes. The installation employs a communication/controller system for automatically identifying a vehicle, and upon authorizing the vehicle, allowing fuel to be dispensed to said vehicle, determining the quantity of fuel dispensed to the vehicle and securing the dispenser when the vehicle leaves a detection zone. The detection system employs RFID tags placed at both the driver side and the passenger side of the vehicle. Generally oppositely oriented antennas define a detection zone for each of the lanes. A light indicates whether the vehicle is authorized for receiving fuel.