Patents by Inventor Vasu Iyengar
Vasu Iyengar has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 10269369Abstract: System of noise reduction for mobile devices includes blind source separator (BSS) and noise suppressor. BSS receives signals from at least two audio pickup channels. BSS includes sound source separator, voice source detector, equalizer, and auto-disabler. Sound source separator generates signals representing first sound source and second sound source based on signals from the first and the second channels. Voice source detector determines whether the signals representing the first and second sound sources are voice signal or noise signal, respectively. Equalizer scales noise signal to match a level of the voice signal, and generates scaled noise signal. Auto-disabler determines whether to disable BSS. Auto-disabler outputs signals from the at least two audio pickup channels when the BSS is disabled and outputs the voice signal and the scaled noise signal when the BSS is not disabled. Noise suppressor generates clean signal based on outputs from auto-disabler. Other embodiments are also described.Type: GrantFiled: May 31, 2017Date of Patent: April 23, 2019Assignee: Apple Inc.Inventors: Nicholas J. Bryan, Vasu Iyengar
-
Publication number: 20190066710Abstract: A method for controlling a speech enhancement process in a far-end device, while engaged in a voice or video telephony communication session over a communication link with a near-end device. A near-end user speech signal is produced, using a microphone to pick up speech of a near-end user, and is analyzed by an automatic speech recognizer (ASR) without being triggered by an ASR trigger phrase or button. The recognized words are compared to a library of phrases to select a matching phrase, where each phrase is associated with a message that represents an audio signal processing operation. The message associated with the matching phrase is sent to the far-end device, which is used to configure the far-end device to adjust the speech enhancement process that produces the far-end speech signal. Other embodiments are also described.Type: ApplicationFiled: August 28, 2017Publication date: February 28, 2019Inventors: Nicholas J. Bryan, Vasu Iyengar, Aram M. Lindahl
-
Patent number: 10176823Abstract: Electronic system for audio noise processing and noise reduction comprises: first and second noise estimators, selector and attenuator. First noise estimator processes first audio signal from voice beamformer (VB) and generate first noise estimate. VB generates first audio signal by beamforming audio signals from first and second audio pick-up channels. Second noise estimator processes first and second audio signal from noise beamformer (NB), in parallel with first noise estimator and generates second noise estimate. NB generates second audio signal by beamforming audio signals from first and second audio pick-up channels. First and second audio signals include frequencies in first and second frequency regions. Selector's output noise estimate may be a) second noise estimate in the first frequency region, and b) first noise estimate in the second frequency region. Attenuator attenuates first audio signal in accordance with output noise estimate. Other embodiments are also described.Type: GrantFiled: May 9, 2014Date of Patent: January 8, 2019Assignee: Apple Inc.Inventors: Sorin V. Dusan, Aram M. Lindahl, Alexander Kanaris, Vasu Iyengar
-
Publication number: 20180367172Abstract: Systems and methods for reducing effects of time-division multiplexing noise in mobile communications devices. When cellular communication with time-division multiplexing is detected, such as Global System for Mobiles (GSM) communication with Time Division Multiple Access (TDMA) protocol, total energy and energy at a repetition frequency of the time division multiplexing is measured in audio signals received from several microphones located in the device. A control signal indicating microphones affected by TDMA noise is provided to signal processing subsystems that receive audio signals from the microphones. A beam former circuit may combine two or more audio signals to produce beam formed signals. The control signal may further indicate beam formed signals affected by TDMA noise based on a ratio of the energy from the repetition frequency to the total energy in the beam formed signals.Type: ApplicationFiled: June 16, 2017Publication date: December 20, 2018Inventors: Ruchir M. Dave, Ashrith Deshpande, Vasu Iyengar
-
Publication number: 20180350381Abstract: System of noise reduction for mobile devices includes blind source separator (BSS) and noise suppressor. BSS receives signals from at least two audio pickup channels. BSS includes sound source separator, voice source detector, equalizer, and auto-disabler. Sound source separator generates signals representing first sound source and second sound source based on signals from the first and the second channels. Voice source detector determines whether the signals representing the first and second sound sources are voice signal or noise signal, respectively. Equalizer scales noise signal to match a level of the voice signal, and generates scaled noise signal. Auto-disabler determines whether to disable BSS. Auto-disabler outputs signals from the at least two audio pickup channels when the BSS is disabled and outputs the voice signal and the scaled noise signal when the BSS is not disabled. Noise suppressor generates clean signal based on outputs from auto-disabler. Other embodiments are also described.Type: ApplicationFiled: May 31, 2017Publication date: December 6, 2018Inventors: Nicholas J. Bryan, Vasu Iyengar
-
Patent number: 10090001Abstract: Method of speech enhancement using Neural Network-based combined signal starts with training neural network offline which includes: (i) exciting at least one accelerometer and at least one microphone using training accelerometer signal and training acoustic signal, respectively. The training accelerometer signal and the training acoustic signal are correlated during clean speech segments. Training neural network offline further includes (ii) selecting speech included in the training accelerometer signal and in the training acoustic signal, and (iii) spatially localizing the speech by setting a weight parameter in the neural network based on the selected speech included in the training accelerometer signal and in the training acoustic signal. The neural network that is trained offline is then used to generate a speech reference signal based on an accelerometer signal from the at least one accelerometer and an acoustic signal received from the at least one microphone. Other embodiments are described.Type: GrantFiled: August 1, 2016Date of Patent: October 2, 2018Assignee: Apple Inc.Inventors: Lalin S. Theverapperuma, Vasu Iyengar, Sarmad Aziz Malik, Raghavendra Prabhu
-
Patent number: 9966067Abstract: Digital signal processing techniques for automatically reducing audible noise from a sound recording that contains speech. A noise suppression system uses two types of noise estimators, including a more aggressive one and less aggressive one. Decisions are made on how to select or combine their outputs into a usable noise estimate in a different speech and noise conditions. A 2-channel noise estimator is described. Other embodiments are also described and claimed.Type: GrantFiled: June 6, 2013Date of Patent: May 8, 2018Assignee: Apple Inc.Inventors: Vasu Iyengar, Sorin V. Dusan
-
Publication number: 20180033449Abstract: Method of speech enhancement using Neural Network-based combined signal starts with training neural network offline which includes: (i) exciting at least one accelerometer and at least one microphone using training accelerometer signal and training acoustic signal, respectively. The training accelerometer signal and the training acoustic signal are correlated during clean speech segments. Training neural network offline further includes (ii) selecting speech included in the training accelerometer signal and in the training acoustic signal, and (iii) spatially localizing the speech by setting a weight parameter in the neural network based on the selected speech included in the training accelerometer signal and in the training acoustic signal. The neural network that is trained offline is then used to generate a speech reference signal based on an accelerometer signal from the at least one accelerometer and an acoustic signal received from the at least one microphone. Other embodiments are described.Type: ApplicationFiled: August 1, 2016Publication date: February 1, 2018Inventors: Lalin S. Theverapperuma, Vasu Iyengar, Sarmad Aziz Malik, Raghavendra Prabhu
-
Publication number: 20180033447Abstract: An audio system has a housing in which are integrated a number of microphones. A programmed processor accesses the microphone signals and produces a number of acoustic pick up beams based groups of microphones, an estimation of voice activity and an estimation of noise characteristics on each beam. Two or more beams including a voice beam that is used to pick up a desired voice and a noise beam that is used to provide information to estimate ambient noise are adaptively selected from among the plurality of beams, based on thresholds for voice separation and thresholds for noise-matching. Other embodiments are also described and claimed.Type: ApplicationFiled: August 1, 2016Publication date: February 1, 2018Inventors: Sean A. Ramprashad, Esge B. Andersen, Joshua D. Atkins, Sorin V. Dusan, Vasu Iyengar, Tarun Pruthi, Lalin S. Theverapperuma
-
Publication number: 20170337932Abstract: An audio system has a housing in which are integrated a number of microphones. A programmed processor accesses the microphone signals and produces a number of acoustic pick up beams. A number of separation values are computed, each being a measure of the difference between strength of a respective beam and strength of a noise reference input signal. One of the beams is selected whose separation value is the largest, and the selected beam is applied to a first input of a two-channel noise suppression process, while the noise reference input signal is applied to the second input of the noise suppression process. Other embodiments are also described and claimed.Type: ApplicationFiled: May 19, 2016Publication date: November 23, 2017Inventors: Vasu Iyengar, Ashrith Deshpande, Aram M. Lindahl
-
Patent number: 9736578Abstract: An orientation detector can have a first microphone, a second microphone, and a reference microphone spaced from the first microphone and the second microphone. An orientation processor can be configured to determine an orientation of the first microphone, the second microphone, or both, relative to a user's mouth based on a comparison of a relative strength of a first signal associated with the first microphone to a relative strength of a second signal associated with the second microphone. A channel selector in a speech enhancer can select one signal from among several signals based at least in part on the orientation determined by the orientation processor. A mobile communication handset can include a microphone-based orientation detector of the type disclosed herein.Type: GrantFiled: June 7, 2015Date of Patent: August 15, 2017Assignee: APPLE INC.Inventors: Vasu Iyengar, Joshua D Atkins, Aram M. Lindahl, Tarun Pruthi, Ashrith Deshpande
-
Patent number: 9524735Abstract: A method for adapting a threshold used in multi-channel audio voice activity detection. Strengths of primary and secondary sound pick up channels are computed. A separation, being a measure of difference between the strengths of the primary and secondary channels, is also computed. An analysis of the peaks in separation is performed, e.g. using a leaky peak capture function that captures a peak in the separation and then decays over time, or using a sliding window min-max detector. A threshold that is to be used in a voice activity detection (VAD) process is adjusted, in accordance with the analysis of the peaks. Other embodiments are also described and claimed.Type: GrantFiled: January 31, 2014Date of Patent: December 20, 2016Assignee: Apple Inc.Inventors: Vasu Iyengar, Aram M. Lindahl
-
Publication number: 20160360314Abstract: An orientation detector can have a first microphone, a second microphone, and a reference microphone spaced from the first microphone and the second microphone. An orientation processor can be configured to determine an orientation of the first microphone, the second microphone, or both, relative to a user's mouth based on a comparison of a relative strength of a first signal associated with the first microphone to a relative strength of a second signal associated with the second microphone. A channel selector in a speech enhancer can select one signal from among several signals based at least in part on the orientation determined by the orientation processor. A mobile communication handset can include a microphone-based orientation detector of the type disclosed herein.Type: ApplicationFiled: June 7, 2015Publication date: December 8, 2016Inventors: Vasu Iyengar, Joshua D. Atkins, Aram M. Lindahl, Tarun Pruthi, Ashrith Deshpande
-
Patent number: 9516159Abstract: System of improving sound quality includes loudspeaker, microphone, accelerometer, acoustic-echo-cancellers (AEC), and double-talk detector (DTD). Loudspeaker outputs loudspeaker signal including downlink audio signal from far-end speaker. Microphone generates microphone uplink signal and receives at least one of: near-end speaker, ambient noise, and loudspeaker signals. Accelerometer generates accelerometer-uplink signal and receives at least one of: near-end speaker, ambient noise, and loudspeaker signals. First AEC receives downlink audio, microphone-uplink and double talk control signals, and generates AEC-microphone linear echo estimate and corrected AEC-microphone uplink signal. Second AEC receives downlink audio, accelerometer uplink and double talk control signals, and generates AEC-accelerometer linear echo estimate and corrected AEC-accelerometer uplink signal.Type: GrantFiled: October 12, 2015Date of Patent: December 6, 2016Assignee: Apple Inc.Inventors: Lalin S. Theverapperuma, Vasu Iyengar, Sean A. Ramprashad
-
Patent number: 9516409Abstract: Systems and methods for controlling echo in audio communications between a near-end system and a far-end system are described. The system and method may intelligently assign a plurality of microphone beams to a limited number of echo cancellers for processing. The microphone beams may be classified based on generated statistics to determine beams of interest (e.g., beams with a high ratio of local-voice to echo). Based on this ranking/classification of microphone beams, beams of greater interest may be assigned to echo cancellers while less important beams may temporally remain unprocessed until these beams become of higher importance/interest. Accordingly, a limited number of echo cancellers may be used to intelligently process a larger number of microphone beams based on interest in the beams and properties of echo cancellation performed for each beam.Type: GrantFiled: May 15, 2015Date of Patent: December 6, 2016Assignee: Apple Inc.Inventors: Sean A. Ramprashad, Martin E. Johnson, Vasu Iyengar, Ronald N. Isaac
-
Patent number: 9516413Abstract: Systems and methods are described for storing and reusing previously generated/calculated acoustic environment data. By reusing acoustic environment data, the systems and methods described herein may avoid the increased overhead in generating/calculating acoustic environment data for a location when this data has already been generated and is likely accurate. In particular, the time and complexity involved in determining reverberation/echo levels, noise levels, and noise types may be avoided when this information is available in storage. This previously stored acoustic environment data may not be limited to data generated/calculated by the same audio device. Instead, in some embodiments an audio device may access a centralized repository to leverage acoustic environment data generated/calculated by other audio devices.Type: GrantFiled: September 30, 2014Date of Patent: December 6, 2016Assignee: Apple Inc.Inventors: Vasu Iyengar, Aram M. Lindahl
-
Patent number: 9491545Abstract: In one embodiment, a process for suppressing reverberation begins with a device of a user obtaining a reverberant speech signal from a voice of the user. The device determines a first estimated reverberation component of the reverberant speech signal. The device generates a first de-reverberated output signal with a first reverberation suppression based on the reverberant speech signal and the first estimated reverberation component. Then, the device generates a second improved reverberation component using the first de-reverberated output signal. The device generates a second de-reverberated output signal with a second reverberation suppression based on the reverberant speech signal and the second improved reverberation component.Type: GrantFiled: May 23, 2014Date of Patent: November 8, 2016Assignee: Apple Inc.Inventors: Vasu Iyengar, Martin E. Johnson, Ronald N. Isaac, Aram M. Lindahl
-
Patent number: 9467779Abstract: Digital signal processing for microphone partial occlusion detection is described. In one embodiment, an electronic system for audio noise processing and for noise reduction, using a plurality of microphones, includes a first noise estimator to process a first audio signal from a first one of the microphones, and generate a first noise estimate. The electronic system also includes a second noise estimator to process the first audio signal, and a second audio signal from a second one of the microphones, in parallel with the first noise estimator, and generate a second noise estimate. A microphone partial occlusion detector determines a low frequency band separation of the first and second audio signals and a high frequency band separation of the first and second audio signals to generate a microphone partial occlusion function that indicates whether one of the microphones is partially occluded.Type: GrantFiled: May 13, 2014Date of Patent: October 11, 2016Assignee: Apple Inc.Inventors: Vasu Iyengar, Fatos Myftari, Sorin V. Dusan, Aram M. Lindahl
-
Publication number: 20160127535Abstract: System of improving sound quality includes loudspeaker, microphone, accelerometer, acoustic-echo-cancellers (AEC), and double-talk detector (DTD). Loudspeaker outputs loudspeaker signal including downlink audio signal from far-end speaker. Microphone generates microphone uplink signal and receives at least one of: near-end speaker, ambient noise, and loudspeaker signals. Accelerometer generates accelerometer-uplink signal and receives at least one of: near-end speaker, ambient noise, and loudspeaker signals. First AEC receives downlink audio, microphone-uplink and double talk control signals, and generates AEC-microphone linear echo estimate and corrected AEC-microphone uplink signal. Second AEC receives downlink audio, accelerometer uplink and double talk control signals, and generates AEC-accelerometer linear echo estimate and corrected AEC-accelerometer uplink signal.Type: ApplicationFiled: October 12, 2015Publication date: May 5, 2016Inventors: Lalin S. Theverapperuma, Vasu Iyengar, Sean A. Ramprashad
-
Publication number: 20150341722Abstract: In one embodiment, a process for suppressing reverberation begins with a device of a user obtaining a reverberant speech signal from a voice of the user. The device determines a first estimated reverberation component of the reverberant speech signal. The device generates a first de-reverberated output signal with a first reverberation suppression based on the reverberant speech signal and the first estimated reverberation component. Then, the device generates a second improved reverberation component using the first de-reverberated output signal. The device generates a second de-reverberated output signal with a second reverberation suppression based on the reverberant speech signal and the second improved reverberation component.Type: ApplicationFiled: May 23, 2014Publication date: November 26, 2015Applicant: Apple Inc.Inventors: Vasu IYENGAR, Martin E. JOHNSON, Ronald N. ISAAC, Aram M. LINDAHL