Sony Corporation

Sony Interactive Entertainment Inc.

Sony Group Corporation

Bibhudendu Mohapatra

Keith Resch

Peter Rae Shintani

Morio Usami

Fred Ansfield

William Clay

Pablo Antonio Espinosa

Jose Omar Gonzalez Hernandez

Gregory Carlsson

James Richard Milne

David Andrew Young

Hyehoon Yi

Tonni Sandager Larsen

Abhishek Pramod Patil

Kissei Matsumoto

Seth Hill

Loren F Pineda

Guru Prashanth Balasubramanian

Lobrenzo Wingo

Dipendu Saha

Max Wu

Travis Charles Baurmann

Kazuyuki Shikama

Peter Andris Vanags

An audio source such as a display device configured to present AV content can present the video and send the audio in different languages to the respective devices of different listeners. For example, a device/TV/source can send audio in different languages to connected headphones/smartglasses with speakers/devices/sink. Furthermore, machine learning may be employed both to recognize listeners and correlate them to likely languages and to mimic voices in the played-back audio. Or, the source AV display device may send language in only the selected language of the display device to each listener device, with each receiving listener device converting the audio to the preferred language of the respective listener on the fly.

Audio video translation into multiple languages for respective listeners

A method of providing a remote-control interface on a smart device for an electronic device by operating an app, developed using machine learning (ML) comprising supervised learning, on the smart device to: use a camera in the smart device to capture an image of the electronic device, with minimal requirements on image quality; operate on the captured image to identify the electronic device in terms of make and model; select from a database of remote-control user interfaces a remote-control user interface corresponding to the identified make and model of the electronic device; and present the selected remote-control user interface on the smart device to a user. In some cases, the captured image is tagged with at least one of date, time, and global location information, allowing further identification in terms of at least one of version ID and region of sale.

Enabling a device with camera and display to control other devices (universal remote control app)

A personal voice model is created using a person's voice, essentially, voice cloning. When user wants to send a message to another person from his mobile phone or similar device, the user types the message and it is converted to a speech message using the voice model created. The speech message is delivered either to a voicemail or another medium which accepts a voice message. The text can be converted to a different language as well and send a voice message in a different language.

Secure text-to-voice messaging

An assembly includes a head mount such as smart glasses wearable on a head of a user. An imager is on the head mount and is configured to generate images of objects. A processor accesses the images responsive to a query and presents images of objects on a display of the head mount to assist the user in identifying, for example, where a lost object was last seen.

Smart glasses lost object assistance

A personalized HRTF is generated from a generic HRTF by having a person don headphones, then select a generic HRTF model from a list. Test signals emulating sounds from a source at a specific virtual speaker location are played on the headphones, and the person indicates where he thinks the emulated sounds are coming from. Any difference between the virtual speaker location and the 'heard' location is used to adjust the parameters (amplitude, phase, and frequency response) of the generic HRTF to render a HRTF personalized to the listening person.

Generating personalized end user head-related transfer function (HRTF) from generic HRTF

A personalized room related transfer function (RRTF) for sound bars and speakers within a specific room is generated without the need for photographs of the listener's head or microphones in a headphone apparatus. A smart device aids the listener in creating his own RRTF by enabling the listener to select a general RRTF model based on the specific sound bar setup (e.g., sound bar; sound bar+subwoofer; sound bar+rear speakers+height; etc.) from a set of general RRTFs downloaded locally or from the cloud. A test signal is played from a specific driver of the sound bar, and the listener identifies a location at which he would like the perceived sound to originate. If the driver location signal and desired location are not the same, the app adjusts parameters and replays the test signal until the driver signal and the desired location are the same.

Generating personalized end user room-related transfer function (RRTF)

Each of plural devices includes a laser emitter and a camera for detecting reflections of laser light emitted by the device, so that plural of the devices can generate their own depth maps showing images within the field of view of their cameras. The resolution of each depth map may be improved by accessing a data store of prior images of objects and combining the prior images using super-resolution with current images of the depth maps.

Using super imposition to render a 3D depth map

A head mounted display (HMD) device that includes a plurality of sensors, an actuator, a plurality of protective mechanisms, and a processor configured to identify a situational state of a user of the HMD device within a proximity of the user, based on sensor input received from the plurality of sensors. A risk-level of a potential injury to the user from a component of the HMD device is determined based on the identified situational state and the sensor input. A protective mechanism is selected from the plurality of protective mechanisms based on the identified situational state and the determined risk-level. The actuator is controlled to deploy the selected protective mechanism to mitigate injury to the user from the component of the HMD device, based on the identified situational state and the determined risk-level.

Protection for head mounted display device

Implementations generally relate to automated equalizer adjustments based on audio characteristics. In some implementations, a method includes detecting music that is being currently played on an audio device. The method further includes adjusting one or more equalizer settings of an equalizer device based at least in part on a music genre associated with the music. The method further includes outputting the music based at least in part on the adjusting of the one or more equalizer settings of the equalizer device.

Adjusting an equalizer based on audio characteristics

Machine learning is applied to both 2D images from an infrared imager imaging laser reflections from an object and to the 3D depth map of the object that is generated using the 2D images and time of flight (TOF) information. In this way, the 3D depth map accuracy can be improved without increasing laser power or using high resolution imagers.

3D depth map

A camera based depth mapping system. Depth information is coded with colors to make a laser-generated 3D depth map easier to interpret. In the event that the laser illumination is not sufficient, the depth information can have a low signal to noise ratio, i.e., the depth map can be noisy. Color noise reduction techniques are used to alleviate this problem.

Color noise reduction in 3D depth map

A micro mirror assembly widens the field of view (FOV) of a 3D depth map sensor, alleviating the limitation of limited laser illumination power and limitations on the resolution of the available imaging device.

Using micro mirrors to improve the field of view of a 3D depth map

Multiple camera systems which use laser illumination for ranging benefit from a heartbeat to synchronize the timing of the individual lasers to eliminate collisions of multiple simultaneous laser emissions. An optical micromesh system provides for the dynamic entry and egress of mobile and stationary nodes. The optical communication between nodes may be controlled using time domain multiplexing (TDM) to avoid cross talk and communication collision due to overlapping laser illumination. By using an optical, laser based micromesh network, laser ranging can be used to dynamically map node location.

Optical micromesh for computerized devices

A head-mounted device (HMD) that includes at least one speaker, at least one audio capturing device and circuitry is provided. The circuitry receives, via the at least one audio capturing device, a user input that corresponds to an audio input. The circuitry further determines a first location of a source of the received audio input, based on the received audio input. The circuitry further determines that the received audio input is enunciated from a wearer of the HMD in a worn state of the HMD, based on the determined first location of the source of the received audio input. The circuitry further controls at least a function of the HMD, based on the determination that the received audio input is enunciated from the wearer.

Head-mounted device control based on wearer information and user inputs

A code is generated by a primary cloud server and sent to a device to be controlled (controlee). The code is displayed and a user speaks the code to a controller device, such as a voice assistant (VA). The controller sends the spoken code via a third party server to the primary server and if the primary server identifies the code as being correct, a first step is complete. The controlee then generates a second code, sends it to the primary server and also displays it for a user to speak it to the controller, which sends the spoken second code to the primary server via the third party server. If the primary server identifies the second code as being correct, pairing is authorized between the controller and controlee. The first and second steps should be completed within a timeout period to be valid.

Growing community of inventors

San Diego, CA, United States of America

Bibhudendu Mohapatra