Psychological counseling auxiliary system based on voice emotion keyword retrieval

Abstract

The invention discloses a psychological counseling auxiliary system based on voice emotion keyword retrieval, which comprises a user module, a data acquisition module, a voice preprocessing module, a voice conversion module, an index establishment module, a feature extraction module, an emotion recognition module, a model establishment module, a keyword corpus, an emotion classification module and a psychological counseling auxiliary module, compared with the prior art, the system has the beneficial effects that by adding the data acquisition module, different emotions and emotional voice audio data of a user can be collected remotely through audio equipment such as a microphone or a mobile phone, so that reasons limited by control conditions can be solved; an index establishment module is added, so that an index network can be constructed by voice recognition results on the basis of keyword retrieval of voice recognition.

Description

Technical Field

[0001] The present invention relates to the technical field of psychological counseling, and specifically to a psychological counseling assistance system based on voice emotion keyword retrieval. Background Art

[0002] Psychological counseling refers to the process of providing psychological assistance to inquirers who have problems in psychological adaptation and seek to solve them by using psychological methods. Those who need to solve problems and come for help are called visitors or clients, and the consulting experts who provide help are called counselors. Visitors describe, inquire, and discuss with counselors about their own psychological discomfort or psychological disorders through communication media such as language and writing. With their support and help, the causes of psychological problems are found through joint discussions, the cruxes of the problems are analyzed, and then conditions and countermeasures for getting out of trouble and solving problems are sought in order to restore psychological balance, improve the adaptability to the environment, and enhance physical and mental health.

[0003] Currently, the existing methods of psychological counseling are generally face-to-face communication. Although this method has more significant effects, it is restricted by too many conditions and it is difficult to provide psychological counseling assistance to more people. Moreover, the existing audio retrieval is basically based on speech recognition technology, which still has certain deficiencies, with too few retrieved keywords and too low recognition rate, and thus it is difficult to support keyword recognition retrieval of voice emotions for a large vocabulary. Summary of the Invention

[0004] The purpose of the present invention is to provide a psychological counseling assistance system based on voice emotion keyword retrieval to solve the problems raised in the above background art, that is, the existing methods of psychological counseling are generally face-to-face communication. Although this method has more significant effects, it is restricted by too many conditions and it is difficult to provide psychological counseling assistance to more people. Moreover, the existing audio retrieval is basically based on speech recognition technology, which still has certain deficiencies, with too few retrieved keywords and too low recognition rate, and thus it is difficult to support keyword recognition retrieval of voice emotions for a large vocabulary.

[0005] To achieve the above object, the present invention provides the following technical solutions: a psychological counseling assistance system based on voice emotion keyword retrieval, including a user module, a data collection module, a voice preprocessing module, a voice conversion module, an index establishment module, a feature extraction module, an emotion recognition module, a model establishment module, a keyword corpus, an emotion classification module, and a psychological counseling assistance module. The output end of the user module is unidirectionally connected to the input end of the data collection module. The output end of the data collection module is unidirectionally connected to the input end of the voice preprocessing module. The output end of the voice preprocessing module is unidirectionally connected to the input end of the voice conversion module. The output end of the voice conversion module is unidirectionally connected to the input end of the index establishment module. The output end of the voice preprocessing module is unidirectionally connected to the input end of the feature extraction module. The output end of the feature extraction module is unidirectionally connected to the input end of the emotion recognition module. The output end of the emotion recognition module is unidirectionally connected to the input end of the emotion classification module. The output end of the emotion classification module is unidirectionally connected to the input end of the model establishment module. Both the index establishment module and the model establishment module are bidirectionally connected to the keyword corpus. The output ends of the index establishment module and the model establishment module are both connected to psychological counseling;

[0006] The user module is used to establish independent user information for convenient recording of user voices and for differentiation;

[0007] The data collection module is used to collect different emotional and emotional voice audio data of users through audio devices such as microphones or mobile phones;

[0008] The voice preprocessing module is used to suppress noise, enhance the audio signal, and supplement audio defects for the collected audio data, so as to improve the audio quality;

[0009] The voice conversion module is used to convert the processed audio data into text;

[0010] The index establishment module is used to construct an index network based on the keyword retrieval of voice recognition for the results of voice recognition;

[0011] The feature extraction module is used to extract features from audio, energy, pitch, speech rate, intonation, and phonemes in voice data for training and recognition;

[0012] The emotion recognition module is used to transmit the voice input to be recognized to a trained emotion recognition model. The model will analyze and process the voice and output the corresponding emotion or emotion category;

[0013] The model establishment module is used to train an emotion recognition model through machine learning or deep learning techniques;

[0014] The keyword corpus is used to store keywords in the corpus through an engine algorithm;

[0015] The emotion classification module is used to classify the emotions of users in detail;

[0016] The psychological counseling assistance module is used to provide psychological counseling assistance based on the retrieval and analysis of the user's voice emotion keywords.

[0017] As a preferred embodiment of the present invention: The data acquisition module includes a recording device unit and an audio storage unit, and the output end of the data acquisition module is unidirectionally connected to the input ends of the recording device unit and the audio storage unit;

[0018] The recording device unit is used to record the user's voice through a recording device such as a microphone or a mobile phone;

[0019] The audio storage unit is used to store the recorded user voice.

[0020] As a preferred embodiment of the present invention: The voice preprocessing module includes a noise suppression unit, a signal enhancement unit, and a defect compensation unit, and the output end of the voice preprocessing module is unidirectionally connected to the input ends of the noise suppression unit, the signal enhancement unit, and the defect compensation unit;

[0021] The noise suppression unit is used to suppress the noise in the audio data;

[0022] The signal enhancement unit is used to enhance the emotion data in the audio data;

[0023] The defect compensation unit is used to compensate for the defects existing in the audio data when suppressing noise.

[0024] As a preferred embodiment of the present invention: The voice conversion module includes a data receiving unit and a voice-to-text unit, and the output end of the voice conversion module is unidirectionally connected to the input ends of the data receiving unit and the voice-to-text unit;

[0025] The data receiving unit is used to receive the preprocessed voice audio data;

[0026] The voice-to-text unit is used to convert the received voice audio data into text;

[0027] The index establishment module includes an index design unit, an index creation unit, and an attribute setting unit, and the output end of the index establishment module is unidirectionally connected to the input ends of the index design unit, the index creation unit, and the attribute setting unit;

[0028] The index design unit is used to determine the columns to be used, select the index type, select appropriate index options, and determine the filegroup or partition scheme layout;

[0029] The index creation unit is used to create an index through SQL;

[0030] The attribute setting unit is used to set attributes such as the name, uniqueness, and sorting method of the index when creating the index.

[0031] As a preferred solution of the present invention: The feature extraction module includes a transformation unit and a feature extraction unit, and the output ends of the feature extraction module are unidirectionally connected to the input ends of the transformation unit and the feature extraction unit;

[0032] The transformation unit is used to extract audio features by analyzing components of different frequencies through the Fourier transform method;

[0033] The feature extraction unit is used to cooperate with the transformation unit to extract the fundamental frequency, energy, tone, speech rate, intonation, and phonemes of the speech.

[0034] As a preferred solution of the present invention: The emotion recognition module includes a data input unit and an emotion recognition unit, and the output ends of the emotion recognition module are unidirectionally connected to the input ends of the data input unit and the emotion recognition unit;

[0035] The data input unit is used to receive audio data and deliver the audio data to the established model for analysis and processing;

[0036] The emotion recognition unit is used to recognize the emotion in the audio data through the established model.

[0037] As a preferred solution of the present invention: The model establishment module includes a data application unit and a model establishment unit, and the output ends of the model establishment module are unidirectionally connected to the input ends of the data application unit and the model establishment unit;

[0038] The data application unit is used to receive the processed data and make reasonable use of it;

[0039] The model establishment unit is used to train an emotion recognition model through machine learning or deep learning techniques using the running data.

[0040] As a preferred solution of the present invention: The keyword corpus includes a data storage unit and a data search unit, and the output ends of the keyword corpus are unidirectionally connected to the input ends of the data storage unit and the data search unit;

[0041] The data storage unit is used to store all keywords for emotion recognition;

[0042] The data search unit is used to facilitate the search for keywords in cooperation with the established model when performing keyword recognition for emotional states.

[0043] As a preferred embodiment of the present invention: The emotion classification module includes an algorithm application unit and a feature classification unit, and the output end of the emotion classification module is unidirectionally connected to the input ends of the algorithm application unit and the feature classification unit;

[0044] The algorithm application unit is used for reasonably applying machine learning algorithms;

[0045] The feature classification unit is used to classify the characteristic emotions of the user through the algorithm application unit.

[0046] As a preferred embodiment of the present invention: The psychological counseling assistance module includes an AI application unit and a psychological counseling unit, and the output end of the psychological counseling assistance module is unidirectionally connected to the input ends of the AI application unit and the psychological counseling unit;

[0047] The AI application unit is used for reasonably applying when analyzing the emotional mood of the user's audio data through AI algorithms;

[0048] The psychological counseling unit is used to cooperate with the AI application unit to provide reasonable psychological counseling to the user and give analysis suggestions.

[0049] Compared with the prior art, the beneficial effects of the present invention are: By adding a data acquisition module, the present invention realizes remotely collecting different emotional and emotional voice audio data of users through audio devices such as microphones or mobile phones, and thus can solve the reason restricted by conditions; By adding an index establishment module, it realizes that the results of speech recognition can be constructed into an index network based on keyword retrieval of speech recognition; By adding a feature extraction module, it realizes extracting features from audio, energy, tone, speech rate, intonation, and phonemes in speech data for training and recognition; By adding an emotion recognition module, it realizes that the speech to be recognized is input and transmitted to the trained emotion recognition model. The model will analyze and process the speech and output the corresponding emotion or mood category; By adding a model establishment module, it realizes training an emotion recognition model through machine learning or deep learning techniques; Furthermore, through the above, the phenomenon of insufficient audio retrieval speech recognition technology is solved, and through the increased keyword corpus, the phenomenon of too few retrieval keywords and too low recognition rate is solved, and it provides support for keyword recognition retrieval of speech emotions with a large vocabulary. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a system block diagram of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0051] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0052] Please refer to Figure 1 , the present invention provides a technical solution: a psychological counseling assistance system based on voice emotion keyword retrieval, including a user module, a data acquisition module, a voice preprocessing module, a voice conversion module, an index establishment module, a feature extraction module, an emotion recognition module, a model establishment module, a keyword corpus, an emotion classification module, and a psychological counseling assistance module. The output end of the user module is unidirectionally connected to the input end of the data acquisition module, the output end of the data acquisition module is unidirectionally connected to the input end of the voice preprocessing module, the output end of the voice preprocessing module is unidirectionally connected to the input end of the voice conversion module, the output end of the voice conversion module is unidirectionally connected to the input end of the index establishment module, the output end of the voice preprocessing module is unidirectionally connected to the input end of the feature extraction module, the output end of the feature extraction module is unidirectionally connected to the input end of the emotion recognition module, the output end of the emotion recognition module is unidirectionally connected to the input end of the emotion classification module, the output end of the emotion classification module is unidirectionally connected to the input end of the model establishment module, both the index establishment module and the model establishment module are bidirectionally connected to the keyword corpus, and the output ends of the index establishment module and the model establishment module are both connected to psychological counseling;

[0053] The user module is used to establish independent user information for convenient recording of user voices and for differentiation;

[0054] The data acquisition module is used to collect different emotional and emotional voice audio data of users through audio devices such as microphones or mobile phones;

[0055] The voice preprocessing module is used to suppress noise, enhance the audio signal, and supplement audio defects for the collected audio data, so as to improve the audio quality;

[0056] The voice conversion module is used to convert the processed audio data into text;

[0057] The index establishment module is used to construct an index network based on the keyword retrieval of voice recognition for the results of voice recognition;

[0058] The feature extraction module is used to extract features from audio, energy, pitch, speech rate, intonation, and phonemes in voice data for training and recognition;

[0059] The emotion recognition module is used to transfer the voice input to be recognized to a trained emotion recognition model. The model analyzes and processes the voice and outputs the corresponding emotion or mood category;

[0060] The model building module is used to train an emotion recognition model through machine learning or deep learning techniques;

[0061] The keyword corpus is used to store keywords in the corpus through an engine algorithm;

[0062] The emotion classification module is used to subdivide and classify the emotions of users;

[0063] The psychological counseling assistance module is used to provide psychological counseling assistance based on the retrieval and analysis of the voice emotion keywords of users.

[0064] Among them, the data acquisition module includes a recording device unit and an audio storage unit, and the output end of the data acquisition module is unidirectionally connected to the input ends of the recording device unit and the audio storage unit;

[0065] The recording device unit is used to record the user's voice through a recording device such as a microphone or a mobile phone;

[0066] The audio storage unit is used to store the recorded user's voice.

[0067] Among them, the voice preprocessing module includes a noise suppression unit, a signal enhancement unit, and a defect compensation unit, and the output end of the voice preprocessing module is unidirectionally connected to the input ends of the noise suppression unit, the signal enhancement unit, and the defect compensation unit;

[0068] The noise suppression unit is used to suppress the noise in the audio data;

[0069] The signal enhancement unit is used to enhance the emotion data in the audio data;

[0070] The defect compensation unit is used to compensate for the defects existing in the audio data when suppressing the noise.

[0071] Among them, the voice conversion module includes a data receiving unit and a voice-to-text unit, and the output end of the voice conversion module is unidirectionally connected to the input ends of the data receiving unit and the voice-to-text unit;

[0072] The data receiving unit is used to receive the voice audio data after the preprocessing is completed;

[0073] The voice-to-text unit is used to convert the received voice audio data into text;

[0074] The index building module includes an index design unit, an index creation unit, and an attribute setting unit, and the output end of the index building module is unidirectionally connected to the input ends of the index design unit, the index creation unit, and the attribute setting unit;

[0075] The index design unit is used to determine the columns to be used, select the index type, select appropriate index options, and determine the filegroup or partition scheme layout;

[0076] The index creation unit is used to create an index through SQL;

[0077] The attribute setting unit is used to set attributes such as the name, uniqueness, sorting method, etc. of the index when creating the index.

[0078] Among them, the feature extraction module includes a transformation unit and a feature extraction unit. The output ends of the feature extraction module are unidirectionally connected to the input ends of the transformation unit and the feature extraction unit;

[0079] The transformation unit is used to extract audio features by analyzing components of different frequencies through the Fourier transform method;

[0080] The feature extraction unit is used to cooperate with the transformation unit to extract the fundamental frequency, energy, tone, speech rate, intonation, and phonemes of speech.

[0081] Among them, the emotion recognition module includes a data input unit and an emotion recognition unit. The output ends of the emotion recognition module are unidirectionally connected to the input ends of the data input unit and the emotion recognition unit;

[0082] The data input unit is used to receive audio data and deliver the audio data to the established model for analysis and processing;

[0083] The emotion recognition unit is used to recognize the emotion in the audio data through the established model.

[0084] Among them, the model establishment module includes a data utilization unit and a model establishment unit. The output ends of the model establishment module are unidirectionally connected to the input ends of the data utilization unit and the model establishment unit;

[0085] The data utilization unit is used to receive the processed data and make reasonable use of it;

[0086] The model establishment unit is used to train the emotion recognition model through the data in operation by machine learning or deep learning techniques.

[0087] Among them, the keyword corpus includes a data storage unit and a data search unit. The output ends of the keyword corpus are unidirectionally connected to the input ends of the data storage unit and the data search unit;

[0088] The data storage unit is used to store all the keywords for emotion recognition;

[0089] The data search unit is used to facilitate the search for keywords in cooperation with the established model when performing keyword recognition of emotional states.

[0090] Among them, the sentiment classification module includes an algorithm application unit and a feature classification unit, and the output ends of the sentiment classification module are unidirectionally connected to the input ends of the algorithm application unit and the feature classification unit;

[0091] The algorithm application unit is used to reasonably utilize machine learning algorithms;

[0092] The feature classification unit is used to classify the feature sentiment of the user through the algorithm application unit.

[0093] Among them, the psychological counseling assistance module includes an AI application unit and a psychological counseling unit, and the output ends of the psychological counseling assistance module are unidirectionally connected to the input ends of the AI application unit and the psychological counseling unit;

[0094] The AI application unit is used to reasonably apply when analyzing the emotional mood of the user's audio data through AI algorithms;

[0095] The psychological counseling unit is used to cooperate with the AI application unit to provide reasonable psychological counseling to the user and give analysis suggestions.

[0096] Specifically, when in use, the independent user information is established through the user module to facilitate recording the user's voice and making distinctions. The data acquisition module collects different emotional and mood voice audio data of the user through audio devices such as microphones or mobile phones. The recording device unit in the data acquisition module records the user's voice through recording devices such as microphones or mobile phones, and then the recorded user voice is stored through the audio storage unit. The voice preprocessing module suppresses noise, enhances the audio signal, and supplements audio defects for the collected audio data to improve the audio quality. The noise suppression unit in the voice preprocessing module suppresses the noise in the audio data, the signal enhancement unit in the voice preprocessing module enhances the emotional data in the audio data, and the defect supplement unit in the voice preprocessing module supplements the defects existing in the audio data during noise suppression. The voice conversion module converts the processed audio data into text. The data receiving unit in the voice conversion module receives the preprocessed voice audio data, and the voice-to-text unit converts the received voice audio data into text. The index establishment module includes an index design unit, an index creation unit, and an attribute setting unit. The index establishment module constructs an index network based on the keyword retrieval of speech recognition for the results of speech recognition. The index design unit in the index establishment module determines the columns to be used, selects the index type, selects appropriate index options, and determines the filegroup or partition scheme layout. The index creation unit creates an index through SQL. When creating an index through the attribute setting unit, attributes such as the name of the index, whether it is unique, and the sorting method are set. The feature extraction module extracts features from the audio, energy, pitch, speech rate, intonation, and phonemes in the voice data for training and recognition. The transformation unit in the feature extraction module extracts audio features by analyzing components of different frequencies through the Fourier transform method. The feature extraction unit cooperates with the transformation unit to extract the fundamental frequency, energy, pitch, speech rate, intonation, and phonemes of the voice. The emotion recognition module inputs the voice to be recognized to the trained emotion recognition model. The model analyzes and processes the voice and outputs the corresponding emotion or mood category. The data input / output unit in the emotion recognition module receives the audio data and transports the audio data to the established model for analysis and processing. The emotion recognition unit recognizes the emotion in the audio data through the established model. The model establishment module trains an emotion recognition model through machine learning or deep learning techniques. The data application unit in the model establishment module receives the processed data and makes reasonable use of it. The model establishment unit trains the emotion recognition model through machine learning or deep learning techniques using the running data. The keyword corpus stores the keywords in the corpus through the engine algorithm. The data storage unit in the keyword corpus stores all the keywords of emotion recognition.When the data search unit identifies emotional keywords, it is convenient to cooperate with the established model to search for keywords. The emotional classification module subdivides and classifies the user's emotions. The algorithm application unit in the emotional classification module reasonably utilizes machine learning algorithms. The feature classification unit classifies the user's characteristic emotions through the algorithm application unit. The psychological counseling assistance module provides psychological counseling assistance based on the retrieval and analysis of the user's voice emotional keywords. The AI application unit in the psychological counseling assistance module reasonably applies AI algorithms when analyzing the emotional state of the user's audio data. The psychological counseling unit cooperates with the AI application unit to provide reasonable psychological counseling to the user and give analysis suggestions.

[0097] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A psychological counseling assistance system based on voice emotion keyword retrieval, characterized by: It includes a user module, a data acquisition module, a speech preprocessing module, a speech conversion module, an index establishment module, a feature extraction module, an emotion recognition module, a model establishment module, a keyword corpus, an emotion classification module, and a psychological counseling auxiliary module, wherein the output end of the user module is unidirectionally connected to the input end of the data acquisition module, the output end of the data acquisition module is unidirectionally connected to the input end of the speech preprocessing module, the output end of the speech preprocessing module is unidirectionally connected to the input end of the speech conversion module, the output end of the speech conversion module is unidirectionally connected to the input end of the index establishment module, the output end of the speech preprocessing module is unidirectionally connected to the input end of the feature extraction module, the output end of the feature extraction module is unidirectionally connected to the input end of the emotion recognition module, the output end of the emotion recognition module is unidirectionally connected to the input end of the emotion classification module, the output end of the emotion classification module is unidirectionally connected to the input end of the model establishment module, the index establishment module and the model establishment module are both bidirectionally connected to the keyword corpus, and the output ends of the index establishment module and the model establishment module are both connected to psychological counseling; The user module is used to establish independent user information to facilitate recording user voices and distinguish them; The data collection module is used to collect different emotional and emotional voice audio data of users through audio devices such as microphones or mobile phones; The speech preprocessing module is used to suppress noise, enhance audio signals and supplement audio defects of the collected audio data, so as to improve the audio quality; The speech conversion module is used to convert the processed audio data into text; The index building module is used to construct the result of speech recognition into an index network based on keyword retrieval of speech recognition; The feature extraction module is used to extract features from audio, energy and tone as well as speech rate, intonation and phonemes in speech data for use in training and recognition; The emotion recognition module is used to pass the speech input to be recognized to the trained emotion recognition model. The model will analyze and process the speech and output the corresponding emotion or mood category; The model building module is used to train an emotion recognition model through machine learning or deep learning technology; The keyword corpus is used to store keywords into the corpus through an engine algorithm; The emotion classification module is used to subdivide and classify the user's emotions; The psychological counseling assistance module is used to provide psychological counseling assistance based on the retrieval and analysis of the user's voice emotion keywords.

2. The psychological counseling assistance system based on voice emotion keyword retrieval according to claim 1 is characterized in that: The data acquisition module comprises a recording device unit and an audio storage unit, and the output end of the data acquisition module is unidirectionally connected to the input end of the recording device unit and the audio storage unit; The recording device unit is used to record the user's voice through a recording device such as a microphone or a mobile phone; The audio storage unit is used to store the recorded user voice.

3. The psychological counseling assistance system based on voice emotion keyword retrieval according to claim 2 is characterized in that: The speech preprocessing module comprises a noise suppression unit, a signal enhancement unit and a defect supplement unit, and the output end of the speech preprocessing module is unidirectionally connected to the input end of the noise suppression unit, the signal enhancement unit and the defect supplement unit; The noise suppression unit is used to suppress noise in the audio data; The signal enhancement unit is used to enhance the emotional data in the audio data; The defect supplementation unit is used to supplement the defects existing in the audio data when suppressing noise.

4. The psychological counseling assistance system based on voice emotion keyword retrieval according to claim 3 is characterized by: The speech conversion module comprises a data receiving unit and a speech-to-text unit, and the output end of the speech conversion module is unidirectionally connected to the input end of the data receiving unit and the speech-to-text unit; The data receiving unit is used to receive the pre-processed voice and audio data; The speech-to-text unit is used to convert the received speech and audio data into text; The index building module includes an index design unit, an index creation unit and an attribute setting unit, and the output end of the index building module is unidirectionally connected to the input end of the index design unit, the index creation unit and the attribute setting unit; The index design unit is used to determine the columns to be used, select the index type, select the appropriate index options, and determine the file group or partition scheme layout; The index creation unit is used to create an index through SQL; The attribute setting unit is used to set the attributes of the index, such as the name, uniqueness, and sorting method, when creating an index.

5. The psychological counseling assistance system based on voice emotion keyword retrieval according to claim 4 is characterized in that: The feature extraction module comprises a transformation unit and a feature extraction unit, and the output end of the feature extraction module is unidirectionally connected to the input end of the transformation unit and the feature extraction unit; The transform unit is used to extract audio features by analyzing components of different frequencies through Fourier transform method; The feature extraction unit is used to cooperate with the transformation unit to extract the fundamental frequency, energy and tone of the speech as well as the speaking speed, intonation and phonemes.

6. The psychological counseling assistance system based on voice emotion keyword retrieval according to claim 5 is characterized by: The emotion recognition module includes a data receiving unit and an emotion recognition unit, and the output end of the emotion recognition module is unidirectionally connected to the input end of the data receiving unit and the emotion recognition unit; The data receiving unit is used to receive audio data and transmit the audio data to the established model for analysis and processing through the model; The emotion recognition unit is used to recognize emotions in audio data through the established model.

7. The psychological counseling assistance system based on voice emotion keyword retrieval according to claim 6 is characterized by: The model building module includes a data application unit and a model building unit, and the output end of the model building module is unidirectionally connected to the input end of the data application unit and the model building unit; The data application unit is used to receive the processed data and make reasonable use of it; The model building unit is used to train the emotion recognition model through running data through machine learning or deep learning technology.

8. The psychological counseling auxiliary system based on voice emotion keyword retrieval according to claim 7 is characterized by: The keyword corpus includes a data storage unit and a data search unit, and the output end of the keyword corpus is unidirectionally connected to the input end of the data storage unit and the data search unit; The data storage unit is used to store all keywords for emotion recognition; The data search unit is used to facilitate searching keywords in conjunction with the established model when performing emotional keyword recognition.

9. The psychological counseling assistance system based on voice emotion keyword retrieval according to claim 8 is characterized by: The emotion classification module includes an algorithm application unit and a feature classification unit, and the output end of the emotion classification module is unidirectionally connected to the input end of the algorithm application unit and the feature classification unit; The algorithm application unit is used to reasonably utilize the machine learning algorithm application; The feature classification unit is used to classify the user's feature emotions through the algorithm application unit.

10. The psychological counseling auxiliary system based on voice emotion keyword retrieval according to claim 9 is characterized in that: The psychological counseling auxiliary module includes an AI application unit and a psychological counseling unit, and the output end of the psychological counseling auxiliary module is unidirectionally connected to the input end of the AI ​​application unit and the psychological counseling unit; The AI ​​application unit is used to reasonably apply the AI ​​algorithm when performing emotional analysis on the user audio data; The psychological counseling unit is used to cooperate with the AI ​​application unit to provide reasonable psychological counseling to users and give analysis suggestions.

Images