Apple Inc.

Dubravko Biruski

Sorin V Dusan

Sungyub D Yoo

Aram M Lindahl

Baptiste Pierre Paquier

A signal processing method to determine whether or not a detected key-phrase is spoken by a wearer of a headphone. The method receives an accelerometer signal from an accelerometer in a headphone and receives a microphone signal from at least one microphone in the headphone. The method detects a key-phrase using the microphone signal and generates a voice activity detection (VAD) signal based on the accelerometer signal. The method determines whether the VAD signal indicates that the detected key-phrase is spoken by a wearer of the headphone. Responsive to determining that the VAD signal indicates that the detected key-phrase is spoken by the wearer of the headphone, triggering a virtual personal assistant (VPA).

Automatic speech recognition imposter rejection on a headphone with an accelerometer

A device implementing an automatic speech recognition triggering system includes at least one processor configured to receive first and second audio signals respectively corresponding to first and second microphones of a device. The at least one processor is further configured to generate, based on at least one of the first or second audio signals, a third audio signal corresponding to a voice beam directed to an expected position of a mouth of a user. The at least one processor is further configured to determine whether wind noise is present in at least one of the first, second, or third audio signals. The at least one processor is further configured to, based on determining whether wind noise is present, an audio signal from among the second or third audio signals, for a determination of whether at least one of the first or second audio signals corresponds to the user.

Noise-dependent audio signal selection system

A digital processor-based memory-efficient and computation-efficient audio signal processing technique partitions each of two audio signals into shorter segments and combines the shorter segments into combined segments. A processor cross-correlates the first combined segment and the second combined segment into a cross-correlation result, which is written into a cross-correlation array. The result may be used for delay estimation (to estimate the relative delay between the two audio signals.) Other aspects are also described and claimed.

Memory and computation efficient cross-correlation and delay estimation

A speech enhancement system for a remote microphone has a wireless receiver that receives a signal from a first microphone of a remote device. A delay buffer receives a second microphone signal from a second microphone and delays by an adjustable delay. The adjustable delay is based on a difference between a wireless delay and an acoustic delay. A noise suppressor produces an output audio signal for an earpiece speaker, based on the first microphone signal and the adjustable delayed second microphone signal. Other aspects are also described and claimed.

Method and system for speech enhancement using a remote microphone

A method performing automatic gain control (AGC) using an accelerometer in a headset starts with an accelerometer-based voice activity detector (VADa) generating a VADa output based on (i) acoustic signals received from at least one microphone included in a pair of earbuds and (ii) data output by at least one accelerometer that is included in the pair of earbuds. The at least one accelerometer detects vibration of the user's vocal chords. The headset includes the pair of earbuds. An AGC controller then performs automatic gain control (AGC) on the acoustic signals from the at least one microphone based on the VADa output. Other embodiments are also described.

Growing community of inventors

Palo Alto, CA, United States of America

Dubravko Biruski