REVIEWS OF ACOUSTICAL PATENTS Free

• JEROME A. HELFFRICH, Southwest Research Institute, San Antonio, Texas 78228

• DAVID PREVES, Starkey Laboratories, 6600 Washington Avenue South, Eden Prairie, Minnesota 55344

• ERIC E. UNGAR, Acentech, Incorporated, 33 Moulton Street, Cambridge, Massachusetts 02138

• JAY KIRSCH, Harman Corporate Technology Group, 10653 South Riverfront Parkway, Suite 300, South Jordan, Utah 84095

Matthieu Taglione and Yannick Caulier, assignors to Areva NP

10 January 2017; filed 8 October 2014

The authors disclose a concept for integrating a camera-based distance measuring unit with an ultrasonic transducer, principally in order to determine the surface profile. They state that this will be done by analyzing the blurring of the images taken, but it is not clear how they deal with near/far ambiguity this way. In one instance of the invention, the camera is movable towards and away by electronic control, allowing the resolution of this ambiguity. Other instances do not have the movable camera. In all cases, the invention is said to result in a more sharply defined ultrasonic image due to the incorporation of the perturbed image. This seems to be less than the current state of the art in ultrasonic inspection.—JAH

Yoshiyuki Abe et al., assignors to NEC Tokin Corporation

10 January 2017; filed 13 July 2012

This patent discloses the construction of a piezoelectric “stack” actuator for use in game controllers, tablet computers, smartphones, and the like. The stack consists of several layers of PZT (lead zirconate titanate) interspersed with conducting foil and nonconducting (plastic) film in such a polarity as to make a “bender” element out of it. These elements contract below the midplane of the actuator and expand above it, as shown in the figure, causing the actuator to bend into an arc. The advantage of this construction is increased displacement of the ends and central region of the surface of the actuator at the expense of lower bandwidth. The authors go on to show the increased output in comparison to other designs, and the effect of encasing the entire actuator in a thin layer of silicone rubber for durability.—JAH

Arthur H. Hartog et al., assignors to Schlumberger Technology Corporation

17 January 2017; filed 19 September 2012

This patent discloses a method for logging a wellbore, for the purpose of locating places where the casing pipe in a well has become disconnected from the cement typically surrounding it. The method disclosed is to use an optical fiber laid in the space around the pipe and then cement the space. When this is done, the fiber becomes sensitive to acoustic strains in the cement, and can be used as a distributed acoustic sensor. By launching an acoustic wave down the pipe and timing its appearance and measuring its attenuation further down the pipe, one can determine the location of the disbond as distance along the pipe.—JAH

Wei Wang et al., assignors to PPG Industries Ohio, Inc

17 January 2017; filed 15 December 2014

This patent describes the chemical composition and processing for coatings to be applied to substrates, such as automobile components, to increase their structural damping and thus improve their sound attenuation performance.—EEU

Donald Wayne Allen et al., assignors to VIV Solutions LLC

17 January 2017; filed 2 June 2015

Vortex-induced vibration is undesirable in tubular structures, such as those employed in underwater drilling and oil production. This patent describes vortex suppression collars that are supported with some gap from a tubular structure by a series of radially extending, circumferentially spaced members. Specific configurations are suggested that can fit into confined spaces, particularly where tubes are joined.—EEU

Rachid Annan, assignor to VSL International AG

24 January 2017; filed 18 December 2014

The device of this patent is intended to provide damping of relative displacements between two elements of civil engineering construction, for example, to reduce the oscillations of the stays of a bridge. The device in essence consists of a series of rods and sleeves that can slide relative to each other and that are pressed against each other by spring arrangements, which can be adjusted to alter the friction between them. Some embodiments include hinges with adjustable friction. Some illustrated configurations are devised to control translational motions and rotations along or about several axes, with adjusted damping of different motions.—EEU

Tetsuya Nobe, assignor to Seiko Instruments Inc

17 January 2017; filed 14 March 2014

This patent describes wristwatch-sized devices that can detect shocks applied to the case and provide a tactile and/or audible signal to the wearer. The device uses piezoelectric elements for sensing and signal generation and includes circuits to provide outputs to display units.—EEU

Kosei Yamashita and Hideyuki Matsunaga, assignors to Sony Corporation

24 January 2017; filed 23 October 2012

The purpose of this patent's device is to provide information about where on a tennis racquet a ball has impacted. A sensor is mounted on the racquet frame and its output is transmitted electronically to a device like a cell phone. The vibration characteristics computed from the sensed vibration spectra are compared with the characteristics corresponding to impacts at each point on the racquet, as stored in memory, and the best fit is chosen as representative of the ball impact location.—EEU

Ching-Pang Lee et al., assignors to Siemens Energy, Inc

17 January 2017; filed 8 July 2010

Helmholtz resonators whose volumes communicate via small openings with areas in which undesirable sound occurs are widely used to attenuate the sound. In high-temperature applications, such as turbine engines on which this patent focuses, cooling of the spaces of concern is a beneficial byproduct of the resonator's action. This patent describes configurations in which added cooling air is admitted through the walls of the resonator by small holes located in the resonator's housing near the surface of the area to be cooled, but at some distance from the openings via which the resonator communicates with the areas of interest. This arrangement provides comparatively efficient cooling without compromising the resonator's damping capability appreciably.—EEU

Jung-Soo Kim and Ik-Soo Kim, assignors to Samsung Electronics Co., Ltd

17 January 2017; filed 14 February 2014

A hearing aid is controlled by an electronic device by means of the following procedure: capturing an image; detecting at least one object in the image that determines a surrounding situation by extracting features of the image; generating a control signal based on the determined surrounding situation; and transmitting the control signal to the hearing aid. Extracted features comprise points formed along a contour of the detected object. The control signal may be used to select a hearing mode suitable for the surrounding situation from several hearing modes that the hearing aid offers. The control signal may comprise at least one of absolute sound level, signal envelope, spectral content, and directional microphone information.—DAP

Adam B. Strauss and Leonid M. Litvak, assignors to Advanced Bionics AG

17 January 2017; filed 23 March 2016

To enhance pitch for cochlear implant wearers, a sound processor identifies at least two frequency bins in an audio signal frequency spectrum presented to a cochlear implant patient that contain spectral energy above and below, respectively, a modified spectral envelope of the audio signal frequency spectrum. The spectral energies in at least two identified frequency bins are enhanced and compressed by applying positive and negative gains, respectively, to the spectral energy in the bins.—DAP

Manuela Feilner and Martin Kuster, assignors to Sonova AG

17 January 2017; filed 24 April 2012

A hearing aid system with left and right hearing devices has two transducers. At least one characteristic feature of received sound information is extracted, from which the type of acoustic environment is determined from several predefined classes based on the extracted characteristic feature(s), and processing parameters are adjusted based on the determined acoustic environment. At least one characteristic feature comprises a complex coherence calculated from the sound information received by the transducers. One transducer may be a pressure microphone and the other may be a particle velocity transducer, both in the same hearing aid in an acoustically coincident manner. The particle velocity transducer may be a pressure gradient microphone or a hot wire particle velocity transducer. Both transducers may be pressure microphones in the same or different hearing aids.—DAP

Roland Barthel and Andre Steinbuss, assignors to Sivantos Pte. Ltd

17 January 2017; filed 8 January 2010

A hearing aid stores and outputs to an external unit the times and listening situations for which at least one predefined signal processing algorithm is activated. In a learning mode, an assessment rating, indicating wearer satisfaction with the processing, is made, inputted, and used to change at least one processing algorithm parameter accordingly. The external unit may be a remote control or personal computer. Situation recognition and/or at least one level meter may be used to determine listening situation.—DAP

Soren Kvist, assignor to GN Resound A/S

24 January 2017; filed 13 June 2013

A transceiver uses an antenna extending on two sides of a behind-the-ear hearing aid, the antenna segment near each side extending to near the opposite side. A feed system excites the antenna to induce a current in at least one segment, the current having first and second local maxima at the first and second sides of the hearing aid, respectively. The antenna may be balanced. One segment may be orthogonal to a surface of the wearer's head. A current bridge may be provided between the two sides of the hearing aid by means of a short circuit from part of the antenna near one side of the hearing aid to part of the antenna near the other side.—DAP

Niels Henrik Pontoppidan et al., assignors to Oticon Medical A/S

31 January 2017; filed 6 July 2015

A method of calibrating a cochlear implant with an electrode array and several sensors includes detecting basilar membrane vibrations and the excitation pattern resulting from an acoustic stimulation provided in the ear canal of a test signal having a known spectrum. The measured excitation pattern of the basilar membrane is compared with an expected excitation pattern based on the acoustic test signal to establish the calibration mapping by assigning characteristic frequencies to the excited positions and the electrode array. The vibration sensor may be an accelerometer or ultrasound sensor that obtains a frequency response. Several sub-sensors may be placed along the length of the cochlear implant to establish a map of the response for each sub-sensor position.—DAP

Thomas Lunner and Niels Henrik Pontoppidan, assignors to Oticon A/S

31 January 2017; filed 11 December 2013

A hearing instrument setting is selected by measuring brain wave signals of a wearer using at least one electrode. The hearing instrument is set into a relaxation mode by a controller if the majority of the measured brain wave signals of the wearer are in the alpha frequency range. The electrodes may be on the surface of the hearing instrument. If the wearer is in an active state, based on the measured brain wave signals, the hearing aid may be set by the controller to an active mode. The relaxation mode setting may comprise at least a reduced gain setting and/or a low pass filter setting. The controller may generate a tinnitus masking sequence when the hearing instrument is set to the relaxation mode.—DAP

Girija Yegnanarayanan et al., assignors to MModal IP LLC

24 January 2017; filed 10 March 2016

To train a speech recognizer, you need, at the very least, a stream of spoken words. However, it also helps to have a transcript to sort out the many ambiguous forms in the spoken materials. In many cases, spoken materials are available, but the only available transcript contains non-literal versions of the spoken materials, synonyms, “corrections,” and the like. This patented system would first attempt to locate “islands,” places in the material where the speech can be aligned with the available text. It would then use general knowledge of the language and of the subject matter to identify concepts and consult various dictionaries to recognize and identify the sections of spoken materials in between the known matches. The exemplary context is patient/doctor interviews.—DLR

Barbara Resch et al., assignors to Dolby International AB

24 January 2017; filed 12 April 2016

Many existing speech analysis systems operate by choosing, moment by moment, whether a given speech data frame is best characterized as voiced or unvoiced and using an appropriate analysis method based on that decision. However, this approach often loses information in the form of non-harmonic content during voiced frames. The present system applies an additional filter following the pitch analysis to attempt to recover such non-harmonic information.—DLR

Francois Mairesse et al., assignors to Amazon Technologies, Inc

31 January 2017; filed 30 March 2015

The first step in speech recognition is the processing of the audio materials to produce a stream of possible identified concepts. An equally important second step is the application of real-world knowledge to connect the identified concepts with specific known referents, persons, objects, processes, and so forth. Often, each of these recognition steps may produce conflicting hypotheses about the intended content. This patent describes several strategies for examining the best results of each technique, combining them into an improved recognition result.—DLR

Dong Yu et al., assignors to Microsoft Technology Licensing, LLC

31 January 2017; filed 8 June 2016

Speech recognition is performed by training first and second neural networks to recognize speech signals spoken by speakers with higher and lower levels, respectively, of a speech characteristic from a mixed speech sample comprising single speaker speech and competing speech. The mixed speech sample is modified to generate multi-condition datasets comprising high and low amplitude characteristic data, respectively. The mixed speech sample is joint decoded with the first and second neural networks using the data for high and low amplitude characteristics, respectively.—DAP

Changsoo Jang, assignor to Changsoo Jang

24 January 2017; filed 26 June 2014

This invention replaces the pole piece wall found in humbucking pickups with an inner coil that is electrically wired to an outer coil. This dual coil system rejects electromagnetic interference like a humbucking pickup, but the coils can be connected in phase to yield the often preferred bright tone of a single coil pickup, or out of phase to yield a warm thick humbucking tone.—JK

Taiji Mizuta et al., assignors to Osaka Alloying Works, Co., Ltd.; Koide Works Limited

17 January 2017; filed 20 November 2015

This alloy strengthens a cymbal by using micronization of the crystal grains by adding zirconium to the typical copper and tin (i.e., bronze) found in cymbals. Increasing the tin proportion improves the sound of a cymbal, but also weakens the cymbal unless a more difficult and expensive manufacturing process is used. Adding zirconium preserves the strength without altering the manufacturing process. Adding small amounts of other elements, such as titanium, silver, or iron, after the zirconium also may improve the sound. Proportions of each of the elements are recommended.—JK

Patrick M. Downey, Bangor, ME

24 January 2017; filed 1 June 2016

This virtual metronome invention allows the user to control the sound of the different beats in a metronome's structured meter. The most important characteristic is space, whereby the sequence of beats can move along a circumference in three-dimensional space on headphones using head-related transforms. This movement of the sound source may include a Doppler shift to make the movement sound more natural. Other characteristics that can be changed by beat include volume, pitch, tone, and timbre. This claims to improve the perception about meter, especially when difficult polymeter and polyrhythms are involved.—JK

Petar Chekardzhikov, Fullerton, CA

17 January 2017; filed 3 December 2014

This invention uses a magnet to affix a tuner to a guitar. This allows the tuner to be hidden from the audience (i.e., to be placed other than on the front of the headstock), allows for easy removal yet longer term mounting, and is less likely to buzz than clip-type mounts when the guitar is played. A swivel can orient the microphone optimally with respect to the instrument.—JK