# US 2026/0247071 A1 · Audio mixing for conference device with multiple microphone arrays

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0247071 A1
- Document ID: US20260247071A1
- Status: Published application
- Kind: A1
- Application number: 19/058,814
- Filed: 2025-02-20
- Published: 2026-08-20
- Earliest claimed: 2025-02-20
- Assignees: Cisco Technology, Inc.
- Applicants: Cisco Technology, Inc.
- Inventors: Asbjorn Therkelsen; Sindre Meringdal; Oivind Stuan
- Web page: https://patentgrep.com/patent/US20260247071A1

## CPC

- H04R 3/005: {for combining the signals of two or more microphones (specially adapted for hearing aids H04R25/407)}
- H04R 1/406: {microphones}
- H04R 27/00: Public address systems (circuits for preventing acoustic reaction H04R3/02; circuits for distributing signals to loudspeakers H04R3/12; {monitoring or testing arrangements for public address systems H04R29/007}; amplifiers H03F)
- G10L 25/21: the extracted parameters being power information
- G10L 25/84: for discriminating voice from noise
- H04R 2201/403: Linear arrays of transducers
- H04R 2410/01: Noise reduction using microphones having different directional characteristics
- H04R 2430/01: Aspects of volume control, not necessarily automatic, in sound systems
- H04R 2430/20: Processing of the output signals of the acoustic transducers of an array for obtaining a desired directivity characteristic (H04R2203/12 takes precedence)
- H04R 2430/23: Direction finding using a sum-delay beam-former
- H04R 2430/25: Array processing for suppression of unwanted side-lobes in directivity characteristics, e.g. a blocking matrix
- H04R: LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; ELECTRIC HEARING AIDS; PUBLIC ADDRESS SYSTEMS (producing sounds with frequency not determined by supply frequency G10K)
- G10L: SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING

## Abstract

A method is performed by a controller of a conference device that includes one or more microphone arrays each to form audio beams to convert audio in a room into beam signals. The method comprises: computing first-stage gains for the beam signals based on bandpass energies of the beam signals; computing second-stage gains for the beam signals based on the first-stage gains and bandpass signal-to-noise ratios of the beam signals; selecting preferred beam signals of the beam signals that have highest second-stage gains; computing transmit gains for the preferred beam signals; weighting the preferred beam signals with the transmit gains, to produce weighted preferred beam signals; and transmitting only the weighted preferred beam signals to a remote conference device.

## Claims

### Claim 1

A method performed by a controller of a conference device that includes one or more microphone arrays each to form audio beams to convert audio in a room into beam signals, the method comprising:
computing first-stage gains for the beam signals based on bandpass energies of the beam signals;
computing second-stage gains for the beam signals based on the first-stage gains and bandpass signal-to-noise ratios of the beam signals;
selecting preferred beam signals of the beam signals that have highest second-stage gains;
computing transmit gains for the preferred beam signals;
weighting the preferred beam signals with the transmit gains, to produce weighted preferred beam signals; and
transmitting only the weighted preferred beam signals to a remote conference device.

### Claim 2 (depends on 1)

The method of claim 1, wherein computing the second-stage gains includes:
estimating total noise energies of the beam signals; and
computing the bandpass signal-to-noise ratios as ratios of the bandpass energies to the total noise energies.

### Claim 3 (depends on 1)

The method of claim 1, further comprising:
detecting a talker in the room and a length of time that the talker has been talking,
wherein computing the second-stage gains includes increasing and decreasing the second-stage gains in correspondence with the length of time the talker has been talking.

### Claim 4 (depends on 1)

The method of claim 1, further comprising:
determining whether the highest second-stage gains are equal,
wherein computing the transmit gains includes computing the transmit gains based on results of determining.

### Claim 5 (depends on 4)

The method of claim 4, further comprising:
setting target gains for the preferred beam signals based on the results of determining,
wherein computing the transmit gains includes computing the transmit gains based on the target gains.

### Claim 6 (depends on 5)

The method of claim 5, wherein:
when the highest second-stage gains are equal, setting the target gains includes setting the target gains equal to each other.

### Claim 7 (depends on 5)

The method of claim 5, wherein:
when the highest second-stage gains are not equal, setting the target gains includes setting a highest target gain and a second highest target gain for a first preferred beam signal and a second preferred beam signal that have a highest second-stage gain and a second highest second-stage gain, respectively.

### Claim 8 (depends on 7)

The method of claim 7, further comprising:
detecting a talker in the room and a length of time that the talker has been talking; and
increasing and decreasing the second highest target gain in correspondence with the length of time the talker has been talking.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
detecting whether voice energy is present in the beam signals;
when the voice energy is present, permitting the transmit gains to increase and decrease over time; and
when the voice energy is not present, only permitting the transmit gains to increase or remain constant over time.

### Claim 10 (depends on 1)

The method of claim 1, wherein:
the beam signals respectively include sequences of audio frames; and
the method further includes computing the first-stage gains, the second-stage gains, and the transmit gains in each audio frame.

### Claim 11 (depends on 10)

The method of claim 10, wherein:
computing the first-stage gains, the second-stage gains, and the transmit gains includes computing the first-stage gains, the second-stage gains, and the transmit gains in a current audio frame based on respective gain results computed in a previous audio frame.

### Claim 12 (depends on 1)

The method of claim 1, wherein:
the beam signals respectively include sequences of audio frames; and
computing each selection gain includes computing a current selection gain for a current audio frame based on a bandpass energy for the current audio frame and a first gain computed for a previous audio frame.

### Claim 13 (depends on 1)

The method of claim 1, wherein the one or more microphone arrays include:
a first microphone array to form first audio beams of the audio beams to produce first beam signals of the beam signals; and
a second microphone array spaced-apart from the first microphone array and used to form second audio beams of the audio beams to produce second beam signals of the beam signals.

### Claim 14

An apparatus comprising:
one or more microphone arrays to form audio beams used to convert audio in a room into beam signals;
a network interface unit to communicate with a network; and
a controller coupled to the one or more microphone arrays and the network interface unit, wherein the controller configured to perform:
computing first-stage gains for the beam signals based on bandpass energies of the beam signals;
computing second-stage gains for the beam signals based on the first-stage gains and bandpass signal-to-noise ratios of the beam signals;
selecting preferred beam signals of the beam signals that have highest second-stage gains;
computing transmit gains for the preferred beam signals;
weighting the preferred beam signals with the transmit gains, to produce weighted preferred beam signals; and
causing only the weighted preferred beam signals to be transmitted to a remote conference device.

### Claim 15 (depends on 14)

The apparatus of claim 14, wherein the controller is configured to perform computing the second-stage gains by:
estimating total noise energies of the beam signals; and
computing the bandpass signal-to-noise ratios as ratios of the bandpass energies to the total noise energies.

### Claim 16 (depends on 14)

The apparatus of claim 14, wherein the controller is further configured to perform:
detecting a talker in the room and a length of time that the talker has been talking,
wherein the controller is configured to perform computing the second-stage gains by increasing and decreasing the second-stage gains in correspondence with the length of time the talker has been talking.

### Claim 17 (depends on 14)

The apparatus of claim 14, wherein the controller is further configured to perform:
determining whether the highest second-stage gains are equal,
wherein computing the transmit gains includes computing the transmit gains based on results of determining.

### Claim 18

A non-transitory computer readable medium encoded with instructions that, when executed by a controller of a conference device that includes one or more microphone arrays each to form audio beams to convert audio in a room into beam signals, cause the controller to perform:
computing first-stage gains for the beam signals based on bandpass energies of the beam signals;
computing second-stage gains for the beam signals based on the first-stage gains and bandpass signal-to-noise ratios of the beam signals;
selecting preferred beam signals of the beam signals that have highest second-stage gains;
computing transmit gains for the preferred beam signals;
weighting the preferred beam signals with the transmit gains, to produce weighted preferred beam signals; and
causing only the weighted preferred beam signals to be transmitted to a remote conference device.

### Claim 19 (depends on 18)

The non-transitory computer readable medium of claim 18, wherein the instructions to cause the controller to perform computing the second-stage gains include instructions to cause the controller to perform:
estimating total noise energies of the beam signals; and
computing the bandpass signal-to-noise ratios as ratios of the bandpass energies to the total noise energies.

### Claim 20 (depends on 18)

The non-transitory computer readable medium of claim 18, further comprising instructions to cause the controller to perform:
determining whether the highest second-stage gains are equal,
wherein the instructions to cause the controller to perform computing the transmit gains include instructions to cause the controller to perform computing the transmit gains based on results of determining.
