US 11,240,601 B1 · Adaptive howling suppressor

Abstract

A system configured to perform adaptive feedback reduction, such as howling suppression, to prevent feedback from occurring when two devices are acoustically coupled during a communication session. To reduce feedback between the acoustically coupled devices, a device monitors input audio data from a microphone and determines whether feedback is present in individual frequency subbands. For example, the device determines an expected value for a first subband using neighboring subbands (e.g., averaging subbands surrounding the first subband) and compares the expected value to an actual value for the first subband. If the actual value is noticeably higher than the expected value, the device determines that feedback is present in the first subband and attenuates the first subband. If the feedback is not present, the device passes the first subband without attenuation.

Claims20

Dependent claims

A computer-implemented method, the method comprising:

receiving first audio data including a first portion corresponding to a first frequency range, a second portion corresponding to a second frequency range higher than the first frequency range, and a third portion corresponding to a third frequency range higher than the second frequency range;
determining a first value associated with the first portion;
determining a second value associated with the second portion;
determining a third value associated with the third portion;
determining, using at least the first value and the third value, a first expected value associated with the second frequency range;
determining, using the first expected value, that the second value satisfies a first condition;
determining a first gain value corresponding to the second frequency range;
and
generating second audio data using the first audio data and the first gain value, the generating comprising attenuating at least a portion of the first audio data.

6 more elements

A system comprising:

at least one processor;
and
memory including instructions operable to be executed by the at least one processor to cause the system to:
receive first audio data including a first portion corresponding to a first frequency range, a second portion corresponding to a second frequency range higher than the first frequency range, and a third portion corresponding to a third frequency range higher than the second frequency range;
determine a first value associated with the first portion;
determine a second value associated with the second portion;
determine a third value associated with the third portion;
determine, using at least the first value and the third value, a first expected value associated with the second frequency range;
determine, using the first expected value, that the second value satisfies a first condition;
determine a first gain value corresponding to the second frequency range;
and
generate second audio data using the first audio data and the first gain value, the generating comprising attenuating at least a portion of the first audio data.

9 more elements

Dependencies are inferred from the claim text.

Adaptive howling suppressor

Life of this patent

Adaptive howling suppressor

Life of this patent

Abstract

Claims20