Method and apparatus for voice perception management in a multi-user environment

Patent metadata

Field	Value
Publication number	US-11978467-B2
Application number	US-202217870759-A
Country	US
Kind code	B2
Filing date	Jul 21, 2022
Priority date	Jul 21, 2022
Publication date	May 7, 2024
Grant date	May 7, 2024

Abstract

A speakerphone includes a processor, a memory device, a power management unit, a first microphone to receive audio waves, a second microphone to receive audio waves, and a third microphone to receive audio waves. The speakerphone may also include a digital signal processor (DSP) to detect a single-user mode activated at the speakerphone, process the audio waves received by the first microphone, second microphone, and third microphone to determine the wave phases of the audio waves received by the first microphone, second microphone, and third microphone, calculate a direction of a voice of a single user relative to the speakerphone; and process the voice of the single user and filter other voices detected by the first microphone, second microphone, and third microphone from the user's voice.

First claim

What is claimed is:

1. A speakerphone comprising:
a memory device;
a power management unit;
a first microphone to receive a first set of audio waves;
a second microphone to receive a second set of audio waves;
a third microphone to receive a third set of audio waves;
a digital signal processor (DSP) to:
detect a single-user mode activated at the speakerphone;
process the audio waves received by the first microphone, second microphone, and third microphone to determine the wave phases of the first set of, second set of, and third set of audio waves received by the first microphone, second microphone, and third microphone, respectively, to determine a single-user direction vector of a single user relative to the speakerphone; and
process the voice of the single user and filter other background voices detected by the first microphone, second microphone, and third microphone from the single user's voice based on background voice direction vectors of the other background voices not matching the single-user direction vector.

2. The speakerphone of claim 1 further comprising:
the DSP further detecting the loudness of the background voices and comparing the loudness of the background voices to the loudness of the single user's voice to distinguish the background voices from the single user's voice.

3. The speakerphone of claim 1 further comprising:
the DSP to monitor for a continuous audio signal of the single user's voice to follow the single user's voice as the user shifts position around the speakerphone.

4. The speakerphone of claim 1 further comprising:
a light-emitting diode (LED) strip indicating an angular field coverage including the direction of where the single user's voice is detected, wherein the indication of the angular field coverage changes as the position of the single user and the detected single user's voice changes.

5. The speakerphone of claim 1 further comprising:
the DSP to detect a loudness level of the single user and compare the loudness of the single user's voice to a loudness threshold over an average duration of time; and
provide feedback to the single user indicating whether the single user's voice is audible or not at the first microphone, second microphone, or third microphone that is a closest microphone to the single user.

6. The speakerphone of claim 1 further comprising:
the DSP to detect a multi-user mode activated at the speakerphone and, with the first microphone, second microphone, and third microphone, process the wave phases of the first set of, second set of, and third set of audio waves received by the first microphone, second microphone, and third microphone, respectively, received from a plurality of users and determining a multi-user direction vector for each user whose voice is above a loudness threshold.

7. The speakerphone of claim 6 further comprising:
a light-emitting diode (LED) strip indicating an angular field coverage indicating the direction of where the plurality of users' voices are detected based on the multi-user vectors for each user.

8. A multi-modal, omnidirectional, speakerphone, comprising:
a memory device;
a power management unit;
a plurality of microphones to receive audio waves;
a digital signal processor (DSP) to detect one of:
a single-user mode activated at the speakerphone; and
a multi-user mode activated at the speakerphone;
the DSP, when the single-user mode is activated, to:
process the audio waves received by the plurality of microphones to determine the wave phases of the audio waves received by each of the plurality of microphones and determine voice characteristics of a single user's voice;
calculate a single-user direction vector of the voice of a single user relative to the speakerphone; and
process the voice of the single user and filter other background voices detected by the plurality of microphones from the single user's voice based on background voice direction vectors of the other background voices not matching the single user direction vector.

9. The multi-modal, omnidirectional, speakerphone of claim 8 further comprising:
the DSP further detecting the loudness of the background voices and comparing the loudness of the background voices to the loudness of the single user's voice to distinguish the background voices from the single user's voice.

10. The multi-modal, omnidirectional, speakerphone of claim 8 further comprising:
the DSP to monitor for a continuous audio signal of the single user's voice to follow the single user's voice as the single user shifts position around the speakerphone.

11. The multi-modal, omnidirectional, speakerphone of claim 8 further comprising:
a light-emitting diode (LED) strip indicating an angular field coverage including the direction of where the single user's voice is detected, wherein the indication of the angular field coverage changes as the position of the single user and the detected single user's voice changes.

12. The multi-modal, omnidirectional, speakerphone of claim 8 further comprising:
the DSP to detect a loudness level of the single user and compare the loudness of the single user's voice to a loudness threshold over an average duration of time; and
provide feedback to the single user indicating whether the user's voice is audible or not at the first microphone, second microphone, or third microphone that is a closest microphone to the single user.

13. The multi-modal, omnidirectional, speakerphone of claim 8 further comprising:
the DSP to detect activation of the multi-user mode activated at the speakerphone and, with the plurality of microphones, process the wave phases of the first set of, second set of, and third set of audio waves received by the first microphone, second microphone, and third microphone, respectively, received from a plurality of users and determining a multi-user direction vectors for each user of the plurality of users.

14. The multi-modal, omnidirectional, speakerphone of claim 13 further comprising:
a light-emitting diode (LED) strip indicating an angular field coverage indicating the direction of where the plurality of users' voices are detected based on the multi-user vectors for each user.

15. A method of operating a speakerphone with a digital signal processor (DSP) comprising:
detecting, with the DSP, a single-user mode activated at the speakerphone;
processing, with the DSP, a first set of audio waves, a second set of audio waves, and a third set of audio waves received by a first microphone, a second microphone, and a third microphone, respectively, to determine the differences in wave phases of the first set of audio waves, the second set of audio waves, and the third set of audio waves to determine a single-user direction vector describing a direction of a voice of a single user relative to the speakerphone; and
processing, with the DSP, the voice of the single user and filtering other background voices detected by the first microphone, second microphone, and third microphone from the single user's voice based on background voice direction vectors of the other background voices not matching the single-user direction vector.

16. The method of claim 15 further comprising:
detecting, using the DSP, the loudness of the background voices and comparing the loudness of the background voices to the loudness of the single user's voice to distinguish the background voices from the single user's voice.

17. The method of claim 15 further comprising:
monitoring, by using the DSP, for a continuous audio signal of the single user's voice to follow the sin

Assignees

• Dell Products Lp

Inventors

• Goh Peng Lip
• Aurongzeb Deeder M
• Chng Eng Kang

Classifications

• G10L21/0216 Primary

  characterised by the method used for estimating noise · CPC title

• G10L25/84

  for discriminating voice from noise · CPC title

• H04R1/08

  Mouthpieces; {Microphones;} Attachments therefor · CPC title

• H04R1/323

  for loudspeakers (H04R1/34 and H04R1/40 take precedence) · CPC title

• G10L2021/02087

  the noise being separate speech, e.g. cocktail party · CPC title