IP Library › Granted Patent US 12,640,027
Granted Patent B2
US 12,640,027 · App. 18/236,236 · Granted May 26, 2026

Systems and methods for increasing the safety of voice conversations between drivers and remote parties

Inventors: Roberto Valter Sicconi (Danbury, CT); Malgorzata Elzbieta Stys (Purdys, NY); Cesar Gonzales (Katonah, NY)
G08B25/00G06N20/00G07C5/008G07C5/085
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 12,640,027
App. No.
18/236,236
Granted
May 26, 2026
Kind
B2
Abstract

A system for increasing the safety of voice conversations between drivers and remote parties, wherein the system includes an in-vehicle subsystem, wherein the in-vehicle subsystem is configured to communicate wirelessly with a remote subsystem accessible to the remote parties in an initial state, a plurality of sensors, and a computing apparatus, wherein the computing apparatus is configured to received monitoring data from the plurality of sensors, determine risk levels as a function of the monitoring data, and alter a state of the communication between the in-vehicle and the remote subsystem as an automatic safety response according to the risk levels, wherein altering the state of the communication includes switching the initial state to a second state as a function of a transition between a first risk level and a second risk level, and transmitting an alert to the remote subsystem as a function of the automatic safety response.

Claims (63)

1 . A system for increasing a safety of voice conversations between drivers and remote parties, wherein the system comprises:

an in-vehicle subsystem, wherein the in-vehicle subsystem is configured to:

communicate wirelessly with a remote subsystem accessible to the remote parties in an initial state;

a plurality of sensors, wherein the plurality of sensors is configured to generate monitoring data; and

a computing apparatus, wherein the computing apparatus is configured to:

receive the monitoring data from the plurality of sensors;

determine a plurality of risk levels comprising a risk level scale as a function of the monitoring data; and

alter a state of the communication between the in-vehicle subsystem and the remote subsystem as an automatic safety response according to the plurality of risk levels, wherein altering the state of the communication comprises:

switching the initial state to a second state as a function of a transition between a first risk level of the plurality of risk levels and a second risk level of the plurality of risk levels;

transmitting an alert to the remote subsystem as a function of the automatic safety response;

operating a dialog interaction engine configured to generate audio prompts to a driver using speech synthesis to drive a speaker array and evaluate responsiveness of the driver using speech recognition and natural language processing, wherein the dialog interaction engine is trained on driver-interaction context data, wherein the evaluated responsiveness recognizes at least a degree of alertness of the driver; and

automatically force a conversation by the remote subsystem to the driver if the degree of alertness is high based on the risk level scale.

2 . The system of claim 1 , wherein the initial state of the communication between the in-vehicle subsystem and the remote subsystem comprises allowing and suspending a remote voice conversation.

3 . The system of claim 2 , wherein altering the state of the communication between the in-vehicle subsystem and the remote subsystem further comprises altering the initial state of the communication using a state machine model, comprising:

activating a digital assistant's dialog flow as a function of the monitoring data; and

switching the second state to a third state as a function of the transition between the second risk level of the plurality of risk levels and a third risk level of the plurality of risk levels.

4 . The system of claim 3 , wherein the third state of the communication between the in-vehicle subsystem and the remote subsystem comprises:

overriding the digital assistant's dialog flow; and

terminating the remote voice conversation.

5 . The system of claim 4 , wherein switching the second state to the third state comprises:

comparing the second risk level of the plurality of risk levels to a second range; and

switching the second state to the third state as a function of the comparison.

6 . The system of claim 1 , wherein the computing apparatus comprises a multimodal neural network, wherein the multimodal neural network is configured to receive as input a combination of text-based digital assistant interactions and a plurality of monitoring data and output textual answers.

7 . The system of claim 1 , wherein switching the initial state to the second state comprises:

comparing the first risk level of the plurality of risk levels to a first range; and

switching the initial state to the second state as a function of the comparison.

8 . The system of claim 1 , wherein determining the plurality of risk levels comprises:

generating a risk machine learning model using training data, wherein the training data comprises a plurality of monitoring data as input correlated to a plurality of risk levels as output; and

determining the plurality of risk levels using the risk machine learning model.

9 . The system of claim 1 , wherein determining the plurality of risk levels further comprises:

transmitting, to the remote subsystem, the monitoring data and an estimated risk level; and

receiving, from the remote subsystem, complementary information to refine and determine a combined risk level.

10 . The system of claim 1 , wherein transmitting the alert comprises notifying emergency services using traffic monitoring services as a function of the alert.

11 . A method for increasing a safety of voice conversations between drivers and remote parties, wherein the method comprises:

generating, by a plurality of sensors, monitoring data;

receiving, by a computing apparatus, the monitoring data from the plurality of sensors;

determining, by the computing apparatus, a plurality of risk levels comprising a risk level scale as a function of the monitoring data;

altering, by the computing apparatus, a state of a communication between an in-vehicle and a remote subsystem as an automatic safety response according to the plurality of risk levels, wherein altering the state of the communication comprises:

switching an initial state to a second state as a function of a transition between a first risk level of the plurality of risk levels and a second risk level of the plurality of risk levels; and

transmitting an alert to the remote subsystem as a function of the automatic safety response;

operating, by the computing apparatus, a dialog interaction engine configured to generate audio prompts to a driver using speech synthesis to drive a speaker array and evaluate responsiveness of the driver using speech recognition and natural language processing, wherein the dialog interaction engine is trained on driver-interaction context data, wherein the evaluated responsiveness recognizes at least a degree of alertness of the driver; and

automatically force a conversation by the remote subsystem to the driver if the degree of alertness is high based on the risk level scale.

12 . The method of claim 11 , wherein the initial state of the communication between the in-vehicle subsystem and the remote subsystem comprises allowing and suspending a remote voice conversation.

13 . The method of claim 12 , wherein altering the state of the communication between the in-vehicle subsystem and the remote subsystem further comprises altering the initial state of the communication using a state machine model, comprising:

activating a digital assistant's dialog flow as a function of the monitoring data; and

switching the second state to a third state as a function of the transition between the second risk level of the plurality of risk levels and a third risk level of the plurality of risk levels.

14 . The method of claim 13 , wherein the third state of the communication between the in-vehicle subsystem and the remote subsystem comprises:

overriding the digital assistant's dialog flow; and

terminating the remote voice conversation.

15 . The method of claim 14 , wherein the computing apparatus comprises a multimodal neural network.

16 . The method of claim 14 , wherein switching the second state to the third state comprises:

comparing the second risk level of the plurality of risk levels to a second range; and

switching the second state to the third state as a function of the comparison.

17 . The method of claim 11 , wherein switching the initial state to the second state comprises:

comparing the first risk level of the plurality of risk levels to a first range; and

switching the initial state to the second state as a function of the comparison.

18 . The method of claim 11 , wherein determining the plurality of risk levels comprises:

generating a risk machine learning model using training data, wherein the training data comprises a plurality of monitoring data as input correlated to a plurality of risk levels as output; and

determining the plurality of risk levels using the risk machine learning model.

19 . The method of claim 11 , wherein determining the plurality of risk levels further comprises:

transmitting, to the remote subsystem, the monitoring data and an estimated risk level; and

receiving, from the remote subsystem, complementary information to refine and determine a combined risk level.

20 . The method of claim 11 , wherein transmitting the alert comprises notifying emergency services using traffic monitoring services as a function of the alert.

Continuity (2)
Continuation In Part 17835010 · Jun 8, 2022
Related Publication 20240112562A1 · Apr 4, 2024
References Cited (30)
US 6731925B2 · Naboulsi · 2004 [cited by examiner]
US 8604932B2 · Breed et al. · 2013 [cited by applicant]
US 9179105B1 · Zeira · 2015 [cited by examiner]
US 9679487B1 · Hayward · 2017 [cited by examiner]
US 9718468B2 · Barfield, Jr. et al. · 2017 [cited by applicant]
US 10235882B1 · Aoude et al. · 2019 [cited by applicant]
US 10247565B2 · Nepomuceno · 2019 [cited by examiner]
US 10268909B2 · Sathyanarayana et al. · 2019 [cited by applicant]
US 10580296B2 · Pedersen · 2020 [cited by examiner]
US 10872379B1 · Nepomuceno · 2020 [cited by examiner]
US 11450206B1 · Fields · 2022 [cited by examiner]
US 20080167820A1 · Oguchi et al. · 2008 [cited by applicant]
US 20130281079A1 · Vidal · 2013 [cited by examiner]
US 20140313208A1 · Filev · 2014 [cited by examiner]
US 20150172450A1 · Singhal · 2015 [cited by examiner]
US 20150186714A1 · Ren · 2015 [cited by applicant]
US 20150294422A1 · Carver · 2015 [cited by examiner]
US 20160236690A1 · Juneja · 2016 [cited by examiner]
US 20170046745A1 · Zhu et al. · 2017 [cited by applicant]
US 20170108864A1 · Wiklinska et al. · 2017 [cited by applicant]
US 20170200061A1 · Julian et al. · 2017 [cited by applicant]
US 20170221149A1 · Hsu-Hoffman · 2017 [cited by examiner]
US 20170292848A1 · Nepomuceno · 2017 [cited by examiner]
US 20180050696A1 · Misu et al. · 2018 [cited by applicant]
US 20180054570A1 · Bai · 2018 [cited by examiner]
US 20180253954A1 · Verma · 2018 [cited by examiner]
US 20190213429A1 · Sicconi · 2019 [cited by examiner]
US 20210270620A1 · Nepomuceno · 2021 [cited by examiner]
Mindtronicai, driver-monitoring-module, Jul. 29, 2019. [cited by applicant]
Charles Murray, NXP, Momenta Partner to Monitor How Alert You Are While Driving, Jun. 28, 2019. [cited by applicant]