IP Library Granted Patent US 10,785,562
Granted Patent B1
US 10,785,562 · App. 16/205,143 · Granted Sep 22, 2020

Position-aware recording devices able to provide context to speech

Inventor: David Emmanuel Weekly (Redwood City, CA)
Assignee: Verily Life Sciences LLC
H04R3/005G10L25/84H04R1/406
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 10,785,562
App. No.
16/205,143
Granted
Sep 22, 2020
Kind
B1
Abstract

Introduced here are recording devices optimized for recording speech between multiple parties (e.g., a speaker and a counterparty, or a first speaker and a second speaker). These recording devices can facilitate the discovery of personalized characteristics of an interaction by detecting location. For example, these recording devices can discover the intent behind certain terms and phrases in the context of a conversation. A recording device can include a communication module configured to stream recorded media content to another electronic device, as well as a marker identification module configured to identify nearby beacons. For example, a recording device may include a Bluetooth® Low Energy chipset configured to identify nearby Bluetooth® beacons and a Wi-Fi® chipset configured to stream recorded media content to another electronic device.

Claims (74)

1. A recording device comprising:

a structural body that includes multiple openings through which acoustic waves corresponding to sounds external to the structural body can travel;

a directional microphone configured to produce first audio data indicative of the sounds external to the structural body,

wherein the directional microphone resides within the structural body adjacent to a first opening of the multiple openings;

an omnidirectional microphone configured to produce second audio data indicative of the sounds external to the structural body,

wherein the omnidirectional microphone resides within the structural body adjacent to a second opening of the multiple openings;

a processing module configured to:

parse the second audio data to discover an environmental noise,

alter the first audio data to mitigate an effect of the environmental noise, and

examine the altered first audio data to identify a conversation involving at least two participants;

an identification module configured to:

monitor for ultrasonic signals that are broadcast by location markers,

wherein each ultrasonic signal includes encoded data that identifies the corresponding location marker responsible for broadcasting the ultrasonic signal, and

generate location data based on a strength of each ultrasonic signal that is presently detectable; and

a communication module configured to:

initiate a communication channel with a server system across a network, and

transmit the altered first audio data and the location data to the server system via the communication channel.

2. A recording device comprising:

a microphone configured to produce audio data indicative of sounds produced by a speaker;

a processing module configured to parse the audio data to identify a conversation involving the speaker and at least one other participant;

an identification module configured to

monitor for signals broadcast by electronic devices located within a specified proximity,

wherein each signal is broadcast by a corresponding electronic device, and

wherein each signal includes data that identifies the corresponding electronic device,

determine that at least one signal is presently being received, and

generate location data based on a strength of the at least one signal; and

a communication module configured to transmit at least some of the audio data, at least some of the location data, or any combination thereof to a destination across a network.

3. The recording device of claim 2 , wherein the microphone is a directional microphone having increased sensitivity to the sounds produced by the speaker.

4. The recording device of claim 2 , wherein the microphone is an omnidirectional microphone designed to record sound with substantially equal gain from all directions.

5. The recording device of claim 2 , wherein the microphone is one of multiple microphones, and wherein the multiple microphones include:

a directional microphone having increased sensitivity to the sounds produced by the speaker, and

an omnidirectional microphone designed to record sound with substantially equal gain from all directions.

6. The recording device of claim 2 , wherein the communication module is further configured to broadcast a reference signal that identifies the recording device.

7. The recording device of claim 2 , wherein each signal of the at least one signal is an ultrasonic signature of a specified frequency.

8. The recording device of claim 2 , wherein the conversation is one of multiple conversations included in the audio data, and wherein the processing module is further configured to:

create a separate file for each conversation of the multiple conversations by splitting the audio data,

wherein each separate file includes a subset of the audio data that is related to the corresponding conversation; and

cause the separate files to be stored in a memory.

9. The recording device of claim 8 , wherein the memory resides on a server system, and wherein said causing comprises:

forwarding the separate files to the communication module for transmission to the server system across the network.

10. The recording device of claim 2 , wherein the processing module is further configured to:

anonymize the audio data by removing or altering an age-specific audible feature, a gender-specific audible feature, or any combination thereof.

11. A method comprising:

acquiring audio data generated by a microphone of a recording device over the course of a conversation involving a speaker;

acquiring location data generated by an identification module of the recording device based on a strength of signals that are detected over the course of the conversation,

wherein each signal is broadcast by a corresponding beacon whose signal can only be detected when located within a specified proximity of the recording device;

temporally aligning the audio data and the location data to establish a location of the speaker over the course of the conversation;

deriving an intent behind an utterance in the conversation by analyzing a segment of audio data that includes the utterance and a segment of location data that temporally corresponds to the segment of audio data; and

performing a specified action based on the intent.

12. The computer-implemented method of claim 11 , wherein the identification module is a Bluetooth Low Energy chipset configured to identify whether a beacon is located within a specified proximity based on whether an identifier broadcast by the beacon is presently being received by the Bluetooth Low Energy chipset.

13. The computer-implemented method of claim 11 , wherein the identification module is a Wi-Fi chipset configured to identify whether a beacon is located within a specified proximity based on whether an identifier broadcast by the beacon is presently being received by the Wi-Fi chipset.

14. The computer-implemented method of claim 11 , further comprising:

applying a speech recognition algorithm to the audio data to create a transcript corresponding to the conversation.

15. The computer-implemented method of claim 11 , wherein said deriving comprises:

identifying a word included in the segment of audio data;

establishing a location of the speaker when the word was spoken by examining the segment of location data; and

predicting the intent based on the word and the location.

16. The computer-implemented method of claim 11 , wherein said performing comprises:

populating a record in a user profile associated with the speaker or another participant in the conversation.

17. The computer-implemented method of claim 11 , wherein said performing comprises:

transmitting an instruction to a display device,

wherein receipt of the instruction prompts the display device to present information related to the speaker or another participant in the conversation.

18. A method comprising:

acquiring, by a processor, audio data generated by a microphone of a recording device over the course of a conversation involving a speaker;

filtering, by the processor, non-ultrasonic frequencies from the audio data;

examining, by the processor, ultrasonic frequencies remaining in the audio data to identify an ultrasonic signature,

wherein the ultrasonic signature is an encoded acoustic signal broadcast by a source;

comparing, by the processor, the ultrasonic signature to a reference database that includes entries for electronic devices arranged throughout an environment,

wherein each entry associates an electronic device with an ultrasonic signature and a location within the environment;

determining, by the processor, that the ultrasonic signature matches a record associated with a given electronic device in the reference database;

establishing, by the processor in response to said determining, that the given electronic device is the source responsible for broadcasting the ultrasonic signature; and

inferring, by the processor, a location of the speaker involved in the conversation based on a known location of the given electronic device associated with as indicated in the matching record in the reference database.

19. The method of claim 18 , wherein the processor resides on the recording device.

20. The method of claim 18 , wherein the processor resides on a network-accessible server system that is communicatively coupled to the recording device.

Assignments (2)
CHANGE OF NAME Recorded Apr 21, 2026
From: VERILY LIFE SCIENCES LLC
To: VERILY HEALTH INC.
Reel/Frame 075477/0981 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jan 9, 2019
From: WEEKLY, DAVID
To: VERILY LIFE SCIENCES LLC
Reel/Frame 047946/0386 →