Back to results
US 12,352,874 B2Granted
Digital monitoring badge system
OwnerVera Kozyr, Dmitrii Lukashev +2inventors
InventorsVera Kozyr, Dmitrii Lukashev, Egor Markov, Igor Mikhnenko
Life of this patent
Estimating the term…
Abstract
A wearable badge for an employee that records and transmits audio from client interactions with the professional, comprising two microphones and two microphone channels that focus one microphone on the speech of the employee and the other microphone on the speech of the customer, making diarizing easier. The wearable badge also comprises a module to determine whether or not the employee is maintaining an appropriate social distance with customers.
Claims20
A system for monitoring customer interactions, comprising:
- at least one badge, wherein each badge is configured to be worn by an employee interacting with a customer, comprising:
- a badge body comprising a front surface, wherein the front surface is configured to display visual information, and a top surface;
- a first microphone and a second microphone located within the badge body;
- a first microphone channel formed within the badge body, wherein the first microphone channel comprises a tubular opening within the badge body leading from the first microphone to the top surface of the badge body;
- a second microphone channel formed within the badge body, wherein the second microphone channel comprises a tubular opening within the badge body leading from the second microphone to the front surface of the badge body;
- a wireless communication module;
- a badge processor and memory, said badge processor and memory configured to perform at least one of the following actions:
- receive audio data from the first microphone and the second microphone;
- store the audio data;
- process the audio data;
- encrypt the audio data;
- transmit the audio data;
- detect voice activity to start recording;
- detect silence to stop recording;
- a hub connected to at least one badge, wherein the hub is a computing device, comprising:
- a communication module;
- a hub processor and memory, said hub processor and memory configured to
- perform the following actions:
- receive audio data from the at least one badge;
- process the audio data by performing at least one of the following actions:
- detect voice activity to delete non-speech parts of the audio;
- detect keywords to timestamp the audio data or to split the audio data;
- diarize the audio data;
- identify age and gender of any speakers;
- identify sounds that may indicate a dangerous incident;
- perform a biometric analysis of any voices to identify at least one person speaking;
- identify any languages spoken in the audio data;
- convert audio data to text;
- send commands to at least one badge.
26 more elements
A method of analyzing an employee's performance, comprising:
- providing a badge to be worn by the employee, comprising a badge body with a front surface and a top surface, a first microphone, and a second microphone, wherein the badge comprises a first microphone channel formed within a badge body, wherein the first microphone channel comprises a tubular opening within the badge body leading from the first microphone to the top surface of the badge body, and a second microphone channel formed within the badge body, wherein the second microphone channel comprises a tubular opening within the badge body leading from the second microphone to the front surface of the badge body;
- using the first microphone and the second microphone to record audio from the employee's interaction with a customer;
- transmitting the audio to a hub, wherein a hub is a computing device;
- using the hub to diarize the audio by performing the following steps:
- analyzing the volume level of audio data received by the first microphone;
- using an AI algorithm to distinguish at least two different voices in the audio data;
- determining the volume level of each voice and determining which voice has the highest volume level;
- erasing all the voices except for the voice with the highest volume level and saving the resulting audio as a first processed audio file;
- analyzing the volume level of audio data received by the second microphone;
- using an AI algorithm to distinguish at least two different voices in the audio data;
- determining the volume level of each voice and determining which voice has the highest volume level;
- erasing all the voices except for the voice with the highest volume level and saving the resulting audio as a second processed audio file;
- using the hub to convert the first processed audio file and the second processed audio file to text;
- using the hub to analyze the text to evaluate the employee's performance.
11 more elements
Dependencies are inferred from the claim text.