Google Inc.

Motorola Corporation

Motorola Mobility LLC

International Business Machines Corporation

Google Technology Holdings LLC

Tenkasi Vaideeswaran Ramabadran

Mark A Jasiuk

Joel A Clark

James Patrick Ashley

Udar Mittal

Michael Joseph McLaughlin

Kevin J Bastyr

Plamen A Ivanov

Alexander Sorin

Jincheng Wu

Giles T Davis

James Brian Piket

William M Kushner

Snehitha Singaraju

Aaron M Smith

Edgardo Manuel Cruz-Zeno

Rivanaldo S Oliveira

Jeffrey Arthur Meunier

Juan C Garcia

Kevin J Bastry

Jeffrey Meunier

A disclosed method includes monitoring an audio signal energy level while having a noise suppressor deactivated to conserve battery power, buffering the audio signal in response to a detected increase in the audio energy level, activating and running a voice activity detector on the audio signal in response to the detected increase in the audio energy level and activating and running a noise estimator in response to voice being detected in the audio signal by the voice activity detector. The method may further include activating and running the noise suppressor only if the noise estimator determines that noise suppression is required. The method activates and runs a noise type classifier to determine the noise type based on information received from the noise estimator and selects a noise suppressor algorithm, from a group of available noise suppressor algorithms, where the selected noise suppressor algorithm is the most power consumption efficient.

Apparatus and method for power efficient signal conditioning for a voice recognition system

An electronic device includes a microphone that receives an audio signal that includes a spoken trigger phrase, and a processor that is electrically coupled to the microphone. The processor measures characteristics of the audio signal, and determines, based on the measured characteristics, whether the spoken trigger phrase is acceptable for trigger phrase model training. If the spoken trigger phrase is determined not to be acceptable for trigger phrase model training, the processor rejects the trigger phrase for trigger phrase model training.

Method and apparatus for evaluating trigger phrase enrollment

An electronic device measures noise variability of background noise present in a sampled audio signal, and determines whether the measured noise variability is higher than a high threshold value or lower than a low threshold value. If the noise variability is determined to be higher than the high threshold value, the device categorizes the background noise as having a high degree of variability. If the noise variability is determined to be lower than the low threshold value, the device categorizes the background noise as having a low degree of variability. The high and low threshold values are between a high boundary point and a low boundary point. The high boundary point is based on an analysis of files including noises that exhibit a high degree of variability, and the low boundary point is based on an analysis of files including noises that exhibit a low degree of variability.

Method and apparatus for estimating variability of background noise for noise suppression

A disclosed method includes monitoring an audio signal energy level while having a plurality of signal processing components deactivated and activating at least one signal processing component in response to a detected change in the audio signal energy level. The method may include activating and running a voice activity detector on the audio signal in response to the detected change where the voice activity detector is the at least one signal processing component. The method may further include activating and running the noise suppressor only if a noise estimator determines that noise suppression is required. The method may activate and runs a noise type classifier to determine the noise type based on information received from the noise estimator and may select a noise suppressor algorithm, from a group of available noise suppressor algorithms, where the selected noise suppressor algorithm is the most power consumption efficient.

One method of operation includes beamforming a plurality of microphone outputs to obtain a plurality of virtual microphone audio channels. Each virtual microphone audio channel corresponds to a beamform. The virtual microphone audio channels include at least one voice channel and at least one noise channel. The method includes performing voice activity detection on the at least one voice channel and adjusting a corresponding voice beamform until voice activity detection indicates that voice is present on the at least one voice channel. Another method beamforms the plurality of microphone outputs to obtain a plurality of virtual microphone audio channels, where each virtual microphone audio channel corresponds to a beamform, and with at least one voice channel and at least one noise channel. The method performs voice recognition on the at least one voice channel and adjusts the corresponding voice beamform to improve a voice recognition confidence metric.

Apparatus and method for beamforming to obtain voice and noise signals

An electronic device includes a microphone that receives an audio signal, and a processor that is electrically coupled to the microphone. The processor detects a trigger phrase in the received audio signal and measure characteristics of the detected trigger phrase. Based on the measured characteristics of the detected trigger phrase, the processor determines whether the detected trigger phrase is valid.

Method and apparatus for mitigating false accepts of trigger phrases

An audio matrix decoder is provided for decoding input signals Land Rinto output signals L, C, S, and R. The decoder circuit includes a main signal path having a passive matrix decoder and a cross talk canceller configured to apply variable gain values to intermediate signals and to compute the output signals L, C, S, and R. The decoder also includes a control signal path having a passive matrix decoder and a linear equation solver. The linear equation solver is configured to compute the variable gain values based on the intermediate signals and the necessary condition that re-encoded values of the output signals L, C, S, and Rare equal to the input signals Land R. The linear equation solver is also configured to feed the variable gain values to the cross talk canceller for use in computing the output signals.

Method and apparatus for audio matrix decoding

An apparatus includes an echo canceller having an audio signal input and an audio signal output and dynamic pre-conditioning logic. The dynamic pre-conditioning logic is operatively coupled to the echo canceller audio signal output as a feedback signal and has a dynamic pre-conditioning logic output operatively coupled to the echo canceller audio signal input. The dynamic pre-conditioning logic is also operative to receive an audio signal input from at least one microphone. The dynamic pre-conditioning logic is operative to analyze the feedback signal to obtain at least one characteristic, and pre-condition the audio signal input, based on the at least one characteristic of the feedback signal, and provide a pre-conditioned audio signal at the echo canceller audio signal input. The echo canceller audio signal output is then provided to a noise suppressor for the send path of a full duplex communication channel.

Apparatus with dynamic audio signal pre-conditioning and methods therefor

Growing community of inventors

Naperville, IL, United States of America

Tenkasi Vaideeswaran Ramabadran