1. A method for analysing and assessing a psychological condition of at least one user by using a machine learning mechanism, comprising: receiving a voice measurement of a duration of to at least an absolute minimum duration; analysing and evaluating the received voice measurement; converting received user voice into a first spectrogram data using discrete wavelet transform (DWT); applying a predicted filter to the first spectrogram data to create an input spectrogram data; inputting the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; and matching at least some of analysis criteria in real-time to assess psychological condition.

2. The method according to claim 1 further comprising step of computing labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.

3. The method according to claim 1 wherein the speech analysis system includes at least a voice analyser processor.

4. The method according to claim 1 wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.

5. The method according to claim 1 wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.

6. A device for analysing and assessing a psychological condition of at least one user comprising: analyse and evaluating the received voice measurement; convert received user voice into a first spectrogram data using discrete wavelet transform (DWT); apply a predicted filter to the first spectrogram data to create an input spectrogram data; input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; match at least some of analysis criteria in real-time to assess psychological condition.

7. The device of claim 6, the device comprises step of compute labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.

8. The device of claim 6, wherein the speech analysis system includes at least a voice analyser processor.

9. The device of claim 6, wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.

10. The device of claim 6, wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.   UTILIZE MACHINE LEARNING BASED VOICE ANALYSIS TO EVALUATE PSYCHOLOGICAL STATE Abstract The present disclosure discloses a device for machine learning based voice analysis to evaluate psychological state. The disclosure includes: analyse and evaluating the received voice measurement; convert received user voice into a first spectrogram data using discrete wavelet transform (DWT); apply a predicted filter to the first spectrogram data to create an input spectrogram data; input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; match at least some of analysis criteria in real-time to assess psychological condition. Fig. XX , Claims:Claims :

1. A method for analysing and assessing a psychological condition of at least one user by using a machine learning mechanism, comprising: receiving a voice measurement of a duration of to at least an absolute minimum duration; analysing and evaluating the received voice measurement; converting received user voice into a first spectrogram data using discrete wavelet transform (DWT); applying a predicted filter to the first spectrogram data to create an input spectrogram data; inputting the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; and matching at least some of analysis criteria in real-time to assess psychological condition.

2. The method according to claim 1 further comprising step of computing labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.

3. The method according to claim 1 wherein the speech analysis system includes at least a voice analyser processor.

4. The method according to claim 1 wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.

5. The method according to claim 1 wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.

6. A device for analysing and assessing a psychological condition of at least one user comprising: analyse and evaluating the received voice measurement; convert received user voice into a first spectrogram data using discrete wavelet transform (DWT); apply a predicted filter to the first spectrogram data to create an input spectrogram data; input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; match at least some of analysis criteria in real-time to assess psychological condition.

7. The device of claim 6, the device comprises step of compute labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.

8. The device of claim 6, wherein the speech analysis system includes at least a voice analyser processor.

9. The device of claim 6, wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.

10. The device of claim 6, wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.

Description:UTILIZE MACHINE LEARNING BASED VOICE ANALYSIS TO EVALUATE PSYCHOLOGICAL STATE
Field of the Invention
[0001] The present invention relates generally to speech analysis, more particularly to a system and method to evaluate a psychological state of a person.
Background
[0002] The background description includes information that may be useful in understanding the present invention. It is not an admission that any of the information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art.
[0003] The ability of the human voice to convey emotions to the listener has long been recognised by scientists. Speech is the acoustic reaction to the vocal cord and vocal tract movements as well as to the resonances of the apertures and cavities of the human head. Muscular tension in the vocal cords and other factors impact how much air comes out of the lungs. This muscular tension is influenced by human emotions as well as some physiological factors that aren't generally connected to the voice, which in turn affects voice modulation. Moreover, certain physiological illnesses like dementia, learning difficulties, and a variety of speech and language impairments with organic roots may also have an impact on how well a person speaks.
[0004] Several speech analysers from the previous art typically rely on one or more extremely specific frequency or time characteristics, or a mix of both, in order to determine the speaker's emotional state. There is no flexibility in the frequency or time domain attributes that are being examined in any of the references.
[0005] Various technological solutions (e.g., mental state analysis apparatus and method using video diary, method of generating relative pattern information between pieces of imitation drawing data, and estimation apparatus and its control method, etc.) are disclosed in patent literature. Few of the exemplary documents are discussed below.
[0006] WO2021261887A1 (By: FLEX CO LTD) relates to a mental state analysis apparatus and method using a video diary, which uses a mental state analysis dedicated app using a video diary performed by a user terminal, wherein, according to the present invention, a service technology environment capable of performing a psychological test by conveniently inducing even a young child who has insufficient reading comprehension capability and writing to express thoughts about a normal life in a video and voice can be provided by the mental state analysis apparatus using a video diary, comprising: a data input unit for receiving an input of video diary data prepared by a video diary preparer via a diary preparation mode of a video diary dedicated application; a mental state analysis unit for analyzing a mental state of the video diary preparer by analyzing the video diary data input to the data input unit; and an analysis result providing unit for providing a result of analyzing the mental state analyzed by the mental state analysis unit via a guardian management mode of the video diary dedicated application.
[0007] US7766828B2 (By: CANON) discloses condition analysis unit (1302) estimates a surrounding environment and psychological state of the user on the basis of image data, voice data, and living body information. When the estimated psychological state is a predetermined state, a cause estimation unit (1303) estimates based on the living body information whether or not the physical condition of the user is bad. When the estimated psychological state is the predetermined state, the cause estimation unit estimates a cause of the psychological state on the basis of the surrounding environment.
[0008] EP3355761A1 (By: CENTRO STUDI) relates to concerns a human emotional/behavioural/psychological state estimation system (1) comprising a group of sensors and devices (11) and a processing unit (12). The group of sensors and devices (11) includes : a video-capture device (111); a skeletal and gesture recognition and tracking device (112); a microphone (113); a proximity sensor (114); a floor pressure sensor (115); user interface means (117); and one or more environmental sensors (118). The processing unit (12) is configured to : acquire or receive a video stream captured by the video-capture device (111) and data items provided by the skeletal and gesture recognition and tracking device (112), the microphone (113), the proximity sensor (114), the floor pressure sensor (115), the environmental sensor (s) (118), and also data items indicative of interactions of a person under analysis with the user interface means (117); detect one or more facial expressions and a position of eye pupils, a body shape and features of voice and breath of the person under analysis; and estimate an emotional/behavioural/'psychological state of said person on the basis of the acquired/received data items, of the detected facial expression (s), position of the eye pupils, body shape and features of the voice and the breath of said person, and of one or more predefined reference mathematical models modelling human emotional/behavioural/psychological states.
[0009] However, the technological solutions for assessing a psychological condition suffers from various limitations such as, inaccuracy, etc. Therefore, more advancement in this field of technology is required. More specifically, to a system and method to evaluate voice based psychological state.
[00010] All references, including publications, patent applications, and patents, cited herein are hereby incorporated by reference to the same extent as if each reference were individually and specifically indicated to be incorporated by reference and were set forth in its entirety herein.

Summary
[00011] The present invention relates generally to speech analysis, more particularly to a system and method to evaluate a psychological state of a person.
[00012] Various objects, features, and advantages of the disclosed subject matter can be more fully appreciated with reference to the following detailed description of the disclosed subject matter when considered in connection with the following drawings, in which like reference numerals identify like elements.
[00013] The following paragraphs provide additional support for the claims of the subject application.
[00014] Embodiments of the present disclosure may include a method for analysing and assessing a psychological condition of at least one user including receiving a voice measurement of a duration of to at least an absolute minimum duration. Embodiments may also include analysing and evaluating the received voice measurement. Embodiments may also include converting received user voice into a first spectrogram data using discrete wavelet transform (DWT).
[00015] Embodiments may also include applying a predicted filter to the first spectrogram data to create an input spectrogram data. Embodiments may also include inputting the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm. Embodiments may also include matching at least some of analysis criteria in real-time to assess psychological condition.
[00016] In some embodiments, the method according to may include step of computing labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels. In some embodiments, the speech analysis system includes at least a voice analyser processor.
[00017] In some embodiments, the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user. Embodiments may also include , if the user's voice may be at least 70 dB louder than the minimum threshold and/or when it may be at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.
[00018] Embodiments of the present disclosure may also include a device for analysing and assessing a psychological condition of at least one user including analyse and evaluating the received voice measurement. Embodiments may also include convert received user voice into a first spectrogram data using discrete wavelet transform (DWT). Embodiments may also include apply a predicted filter to the first spectrogram data to create an input spectrogram data. Embodiments may also include input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm. Embodiments may also include match at least some of analysis criteria in real-time to assess psychological condition.
[00019] In some embodiments, the one or more processors may be further configured to step of compute labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels. In some embodiments, the speech analysis system includes at least a voice analyser processor.
[00020] In some embodiments, the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user. Embodiments may also include, if the user's voice may be at least 70 dB louder than the minimum threshold and/or when it may be at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.
Brief Description of the Drawings
[00021] The features and advantages of the present disclosure would be more clearly understood from the following description taken in conjunction with the accompanying drawings in which:
[00022] FIG. 1 is a flowchart illustrating method for analysing voice to evaluate psychological state, according to some embodiments of the present disclosure.
[00023] FIG. 2 is a block diagram illustrating a device for voice analysis to evaluate psychological state, according to some embodiments of the present disclosure.

Detailed Description
[00024] The following is a detailed description of exemplary embodiments to illustrate the principles of the invention. The embodiments are provided to illustrate aspects of the invention, but the invention is not limited to any embodiment. The scope of the invention encompasses numerous alternatives, modifications and equivalent; it is limited only by the claims.
[00025] In view of the many possible embodiments to which the principles of the present discussion may be applied, it should be recognized that the embodiments described herein with respect to the drawing figures are meant to be illustrative only and should not be taken as limiting the scope of the claims. Therefore, the techniques as described herein contemplate all such embodiments as may come within the scope of the following claims and equivalents thereof.
[00026] Following are more detailed descriptions of various concepts related to, and implementations of, methods, apparatuses, and systems of present disclosure. The various concepts introduced above and discussed in greater detail below may be implemented in any of numerous ways.
[00027] The detailed description is described with reference to the accompanying figures. In the figures, the left-most digit(s) of a reference number identifies the figure in which the reference number first appears. The use of the same reference numbers in different instances in the description and the figures may indicate similar or identical items.
[00028] The present invention relates generally to speech analysis, more particularly to a system and method to evaluate a psychological state of a person.
[00029] FIG. 1 is a flowchart illustrating method for analysing voice to evaluate psychological state, according to some embodiments of the present disclosure. In some embodiments, at 160, the method may include receiving a voice measurement of a duration of to at least an absolute minimum duration. At 110, the receiving may include analysing and evaluating the received voice measurement. At 120, the receiving may include converting received user voice into a first spectrogram data using discrete wavelet transform (DWT). At 130, the receiving may include applying a predicted filter to the first spectrogram data to create an input spectrogram data. At 140, the receiving may include inputting the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm. At 150, the receiving may include matching at least some of analysis criteria in real-time to assess psychological condition.
[00030] In some embodiments, the method may include step of computing labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels. In some embodiments, the speech analysis system may include at least a voice analyser processor. In some embodiments, the user's voice analysis algorithms may employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user. In some embodiments, if the user's voice may be at least 70 dB louder than the minimum threshold and/or when it may be at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.
[00031] FIG. 2 is a block diagram illustrating a device for voice analysis to evaluate psychological state, according to some embodiments of the present disclosure. In some embodiments, the device 200 may include analyse 210 and evaluating the received voice measurement. Convert received user voice into a first spectrogram data using discrete wavelet transform (DWT). Apply a predicted filter to the first spectrogram data to create an input spectrogram data. Input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm. Match at least some of analysis criteria in real-time to assess psychological condition.
[00032] In some embodiments, the one or more processors may be further configured to execute step to compute labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels. The user's voice analysis algorithms may employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user. In some embodiments, if the user's voice may be at least 70 dB louder than the minimum threshold and/or when it may be at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.
[00033] A user terminal may be used in accordance with embodiments of the present disclosure to receive a psychological test start input in relation to a counselling receiver, and utilise this input to determine the recipient's mental health. In certain implementations, the user terminal may be used to designate a space for the counselee to draw in when the user initiates the psychological exam.
[00034] At least one identity-invariant characteristic of a drawn object may be extracted in certain embodiments. In certain implementations, the identity-invariant characteristics are used to create a reference training data set using a first neural network engine. In other implementations, a second neural network is used to determine the sequence in which the guidance receiver draws the region in order to identify one or more bits of instruction information. A psychological state of the counselee might be determined in certain embodiments using an AI-based method.
[00035] In certain implementations, the user terminal will also communicate analytical remarks about the psychologist to a management server. After receiving the psychological test start input, there may be many pause locations available to the counselling receiver in relation to the zone to be drawn in. The strategy may, in certain implementations, include giving the receiver improvised input depending on their assessed mental state. Reading a book, doing yoga, engaging in social engagement, or a mix of these is included in certain implementations of the improvisation feedback.
[00036] A system for determining a psychological state of a counselling recipient may also be included in embodiments of the present disclosure, and may include receiving, through a user terminal, a psychological test start input with regard to a counselling recipient. In certain implementations, the user terminal will prompt the test receiver to draw a certain region after receiving the psychological test start input.
[00037] At least one identity-invariant characteristic of a drawn object may be extracted in certain embodiments. A reference training data set based on the identity-invariant characteristics may be generated by a first neural network engine in certain embodiments. Other embodiments use the use of a second neural network to determine the sequence in which the counselee draws the region in order to identify one or more bits of instruction information. The psychological condition of a counselee might be determined in certain embodiments using an AI-based method.
[00038] In certain implementations, the user terminal will also communicate analytical remarks about the psychologist to a management server. After receiving the psychological test start input, there may be many pause locations available to the counselling receiver in relation to the zone to be drawn in. Other implementations of the method may include giving the receiver feedback on their improv skills depending on their perceived mental state. Improvisational feedback may take the form of reading a book, doing yoga, engaging in a social activity, or a mix of these.
[00039] Variations of those preferred embodiments may become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventors expect skilled artisans to employ such variations as appropriate, and the inventors intend for the invention to be practiced otherwise than as specifically described herein. Accordingly, this invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described elements in all possible variations thereof is encompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context.
[00040] The wordings such as include, including, comprise and comprising do not exclude elements or steps which are present but not listed in the description and the claims.
[00041] It also shall be noted that as used herein and in the appended claims, the singular forms a, an, and the include plural referents unless the context clearly dictates otherwise. This invention can be achieved by means of hardware including several different elements or by means of a suitably programmed computer. In the unit claims that list several means, several ones among these means can be specifically embodied in the same hardware item. The use of such words as first, second, third does not represent any order, which can be simply explained as names.
[00042] Various techniques may be described herein in the general context of software, hardware elements, or program modules. Generally, such modules include routines, programs, objects, elements, components, data structures, and so forth that perform particular tasks or implement particular abstract data types. The terms module, functionality, and component as used herein generally represent software, firmware, hardware, or a combination thereof. The features of the techniques described herein are platform-independent, meaning that the techniques may be implemented on a variety of commercial computing platforms having a variety of processors.

Claims
I/We Claim:
1. A method for analysing and assessing a psychological condition of at least one user by using a machine learning mechanism, comprising: receiving a voice measurement of a duration of to at least an absolute minimum duration; analysing and evaluating the received voice measurement; converting received user voice into a first spectrogram data using discrete wavelet transform (DWT); applying a predicted filter to the first spectrogram data to create an input spectrogram data; inputting the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; and matching at least some of analysis criteria in real-time to assess psychological condition.
2. The method according to claim 1 further comprising step of computing labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.
3. The method according to claim 1 wherein the speech analysis system includes at least a voice analyser processor.
4. The method according to claim 1 wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.

5. The method according to claim 1 wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.
6. A device for analysing and assessing a psychological condition of at least one user comprising:
analyse and evaluating the received voice measurement;
convert received user voice into a first spectrogram data using discrete wavelet transform (DWT);
apply a predicted filter to the first spectrogram data to create an input spectrogram data;
input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm;
match at least some of analysis criteria in real-time to assess psychological condition.
7. The device of claim 6, the device comprises step of compute labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.
8. The device of claim 6, wherein the speech analysis system includes at least a voice analyser processor.
9. The device of claim 6, wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.
10. The device of claim 6, wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.
 

UTILIZE MACHINE LEARNING BASED VOICE ANALYSIS TO EVALUATE PSYCHOLOGICAL STATE
Abstract
The present disclosure discloses a device for machine learning based voice analysis to evaluate psychological state. The disclosure includes: analyse and evaluating the received voice measurement; convert received user voice into a first spectrogram data using discrete wavelet transform (DWT); apply a predicted filter to the first spectrogram data to create an input spectrogram data; input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; match at least some of analysis criteria in real-time to assess psychological condition.

Fig. XX
, Claims:Claims
I/We Claim:
1. A method for analysing and assessing a psychological condition of at least one user by using a machine learning mechanism, comprising: receiving a voice measurement of a duration of to at least an absolute minimum duration; analysing and evaluating the received voice measurement; converting received user voice into a first spectrogram data using discrete wavelet transform (DWT); applying a predicted filter to the first spectrogram data to create an input spectrogram data; inputting the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm; and matching at least some of analysis criteria in real-time to assess psychological condition.
2. The method according to claim 1 further comprising step of computing labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.
3. The method according to claim 1 wherein the speech analysis system includes at least a voice analyser processor.
4. The method according to claim 1 wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.

5. The method according to claim 1 wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.
6. A device for analysing and assessing a psychological condition of at least one user comprising:
analyse and evaluating the received voice measurement;
convert received user voice into a first spectrogram data using discrete wavelet transform (DWT);
apply a predicted filter to the first spectrogram data to create an input spectrogram data;
input the input spectrogram data to a trained machine learning model through an artificial intelligence algorithm;
match at least some of analysis criteria in real-time to assess psychological condition.
7. The device of claim 6, the device comprises step of compute labels for a plurality of output layers corresponding to the deep neural network and training the deep neural network using the input spectrogram data and the computed labels.
8. The device of claim 6, wherein the speech analysis system includes at least a voice analyser processor.
9. The device of claim 6, wherein the user's voice analysis algorithms employ at least resonant neural components to compensate for losses from distortion caused by natural resonances so as to predict noises in a system while interacting with the user.
10. The device of claim 6, wherein, if the user's voice is at least 70 dB louder than the minimum threshold and/or when it is at least 10 dB louder than the minimum threshold, then each speech measurement will be recorded.