Systems and methods for detecting impairment based upon voice data
Various embodiments provide systems and methods for identifying impairment using measurement devices and trained models, and/or for indicating interference with impairment testing. Embodiments discussed herein provide systems, methods, and/or devices that enable remote impairment testing that does not require a human monitor to be present or physically near the individual being monitored. Such an ability is an improvement.
1 . A system for detecting impairment based upon voice data the system comprising:
a microphone configured to receive audio information from an individual reading a message transmitted by a central monitoring station and to provide recorded voice data corresponding to the audio information;
one or more processors;
a non-transient computer readable medium coupled to the one or more processors, and having stored therein instructions which when executed by the one or more processors, causes the one or more processors to:
receive, by an integrated circuit comprising user identification circuitry, the recorded voice data from the microphone;
process the recorded voice data by a voice impairment model to yield a probability that the individual is impaired;
transmit, by transceiver circuitry, an indication of a likelihood of impairment to the central monitoring station, based at least in part on a determination that the probability exceeds a first threshold;
transmit, by the transceiver circuitry, an indication of no impairment to the central monitoring station when the probability is less than a second threshold;
cause a request to be sent to the individual to perform a movement based impairment test when the probability is less than the first threshold but more than the second threshold;
receive movement information from a movement sensor; and
apply a movement impairment model to the movement information,
wherein the movement impairment model indicates a likelihood that the individual is impaired based upon a movement of the individual when they are expected to be standing still.
2 . The system of claim 1 , wherein the non-transient computer readable medium further having stored therein instructions which when executed by the one or more processors, causes the one or more processors to:
apply an anomaly detection model to the recorded voice data to yield an individual anomaly output; and
wherein the likelihood of impairment is indicated when both the individual anomaly output indicates that the recorded voice data is an anomaly for the individual and the probability exceeds the first threshold.
3 . The system of claim 2 , wherein the anomaly detection model is trained using at least ten instances of recorded voice data derived from the individual.
4 . The system of claim 1 , wherein the voice impairment model is a machine learning model trained using at least one hundred instances of recorded voice data.
5 . The system of claim 4 , wherein the at least one hundred instances of recorded voice data correspond to at least ten different individuals undergoing a voice based impairment test.
6 . The system of claim 1 , wherein the non-transient computer readable medium further having stored therein instructions which when executed by the one or more processors, causes the one or more processors to:
cause a request to be sent to the individual to perform an additional impairment test.
7 . The system of claim 6 , wherein the additional impairment test is a facial image based impairment test.
8 . The system of claim 1 , wherein the non-transient computer readable medium further having stored therein instructions which when executed by the one or more processors, causes the one or more processors to:
forward the recorded voice data to a user for classification when the probability is both less than the first threshold and greater than the second threshold.
9 . The system of claim 1 , wherein the non-transient computer readable medium further having stored therein instructions which when executed by the one or more processors, causes the one or more processors to:
report the likelihood of impairment to a recipient device apart from the one or more processors.
10 . A method for detecting impairment based upon voice data, the method comprising:
receiving, by a processor operating an integrated circuit comprising user identification circuitry, recorded voice data captured by a microphone of an individual reading a message transmitted by a central monitoring station;
processing, by the processor, the recorded voice data using a voice impairment model to yield a probability that the individual is impaired;
transmit, by transceiver circuitry, an indication of a likelihood of impairment to the central monitoring station, based at least in part on a determination that the probability exceeds a first threshold;
transmit, by the transceiver circuitry, an indication of no impairment to the central monitoring station when the probability is less than a second threshold;
cause, by the processor, a request to be sent to the individual to perform a movement based impairment test when the probability is less than the first threshold but more than the second threshold;
receive, by the central monitoring station, movement information from a movement sensor; and
apply, by the processor, a movement impairment model to the movement information,
wherein the movement impairment model indicates a likelihood that the individual is impaired based upon a movement of the individual when they are expected to be standing still.
11 . The method of claim 10 , the method further comprising:
applying, by the processor, an anomaly detection model to the recorded voice data to yield an individual anomaly output; and
wherein the likelihood of impairment is indicated when both the individual anomaly output indicates that the recorded voice data is an anomaly for the individual and the probability exceeds the first threshold.
12 . The method of claim 11 , wherein the anomaly detection model is trained using at least ten instances of recorded voice data derived from the individual.
13 . The method of claim 10 , wherein the voice impairment model is a machine learning model trained using at least one hundred instances of recorded voice data.
14 . The method of claim 13 , wherein the at least one hundred instances of recorded voice data correspond to at least ten different individuals undergoing a voice based impairment test.
15 . The method of claim 10 , the method further comprising:
causing, by the processor, a request to be sent to the individual to perform an additional impairment test.
16 . The method of claim 15 , wherein the additional impairment test is a facial image based impairment test.
17 . The method of claim 10 , the method further comprising:
forwarding, by the processor, the recorded voice data to a user for classification when the probability is both less than the first threshold and greater than the second threshold.
18 . The method of claim 10 , the method further comprising:
reporting, by the processor, the likelihood of impairment to a recipient device.
19 . A non-transient computer readable medium having stored therein instructions, which when executed by a hardware processing system cause the hardware processing system to:
receive, by an integrated circuit comprising user identification circuitry a recorded voice data from a microphone, wherein the recorded voice data corresponds to a voice of an individual reading a message transmitted by a central monitoring station;
process the recorded voice data by a voice impairment model to yield a probability that the individual is impaired, wherein the voice impairment model is a machine learning model trained using at least one hundred instances of recorded voice data, and wherein the at least one hundred instances of recorded voice data correspond to at least ten different individuals undergoing a voice based impairment test;
transmit, by transceiver circuitry, an indication of a likelihood of impairment to the central monitoring station, based at least in part on a determination that the probability exceeds a first threshold;
transmit, by the transceiver circuitry, an indication of no impairment to the central monitoring station when the probability is less than a second threshold;
cause a request to be sent to the individual to perform a movement based impairment test when the probability is less than the first threshold but more than the second threshold;
receive movement information from a movement sensor; and
apply a movement impairment model to the movement information,
wherein the movement impairment model indicates a likelihood that the individual is impaired based upon a movement of the individual when they are expected to be standing still.
20 . The non-transient computer readable medium of claim 19 , wherein the non-transient computer readable medium further having stored therein instructions which when executed by the hardware processing system, causes the hardware processing system to:
apply an anomaly detection model to the recorded voice data to yield an individual anomaly output, wherein the anomaly detection model is trained using at least ten instances of voice data derived from the individual; and
wherein the likelihood of impairment is indicated when both the individual anomaly output indicates that the recorded voice data is an anomaly for the individual and the probability exceeds the first threshold.