TW202008384A - Automatic auditory perception evaluation method including a preparation step, an evaluation step and a calculation step - Google Patents

Abstract

This invention discloses an automatic auditory perception evaluation method which is suitable for testing a testee and includes a preparation step, an evaluation step, and a calculation step. In the preparation step, an evaluation apparatus storing a plurality of first test questions is provided, each first test question has a first sound and a plurality of first picture galleries each with a plurality of first pictures. In the evaluation step, the evaluation apparatus selects the first test questions and plays the first sound, then, one of the first pictures is selected from the first picture galleries and is simultaneously displayed, the testee selects one of the first pictures according to the first sound to generate a first answer content. In the calculation step, a first standard answer corresponding to the first test question is compared with the first answer content to generate a first result. Therefore, a relatively accurate evaluation result may be obtained without applying the speaking ability of the testee.

Description

Automatic hearing perception evaluation method and device

The invention relates to an evaluation method and an evaluation device, in particular to an automated hearing perception evaluation method and device for evaluating the perception conceptual ability of patients with neurological brain injury.

Conceptual ability is a dynamic effect established by perception and action experience. It is an important part of human cognition, and it is the basis for language, object recognition, action planning and thinking. Clinical therapists and researchers have always attached great importance to the evaluation and treatment of neurological brain injury. Many patients with neurological brain injury often have different degrees of cognitive impairment. Understanding the conceptual ability of patients with neurological brain injury will help to clarify the disease and benefit The follow-up treatment is carried out. Therefore, by understanding the cognitive level of patients with neurological brain injury, knowing the patient's conceptual ability is an important step before treatment.

The clinically available tools for evaluating patients with neurological brain injury still use English as the main language, which is less suitable for patients who use Chinese or Taiwanese as their common language. In recent years, a Chinese-based scholar from Hong Kong has published a Taiwan version of the assessment tool for conceptual abilities of neurologically impaired patients-The Main Concept Analysis for oral discourse production (MCA for short), which can be used to analyze the oral expression of subjects Learn about its conceptual capabilities.

However, because conceptual abilities are related to language abilities, if a subject has a language impairment due to nerve damage or other factors, using a language-based assessment tool such as MCA to assess their conceptual abilities may be due to the subject It cannot be expressed in spoken language, and it is difficult to obtain accurate assessment results. In other words, MCA is only suitable for patients with oral expression ability, and it is difficult to apply to patients with no oral ability or extremely poor oral expression (such as severe aphasia patients, etc.), and patients who are unable to express oral expression, their perceptual conceptual ability is still Important information that must be obtained before treatment. Therefore, a set of clinical ability assessment tools that can control or eliminate the need for oral expression ability and can be tested only with auditory stimulation is needed clinically.

Therefore, the purpose of the present invention is to provide an automated auditory perception evaluation method that uses a nonverbal way to evaluate the perception conceptual ability.

Therefore, the automated auditory perception evaluation method of the present invention is suitable for providing a subject to be tested, and includes a preparation step, an evaluation step, and a calculation step.

The preparation step provides an evaluation device, which includes a server and a user terminal connected to the server. The server includes a data module storing a plurality of first test questions, a processing module with information connected to the data module, and a storage module with information connected to the processing module. The user terminal includes a broadcasting module with information connected to the data module, a display module with information connected to the data module and the processing module, and an input module with information connected to the display module and the processing module . The first test question of each question has a first sound, and a plurality of first image libraries related to the first sound and having a plurality of first pictures, respectively.

The evaluation step includes a first quantum evaluation step. In the first quantum evaluation step, the server selects a first test question from the data module and transmits it to the user end for the subject to answer. The user terminal plays the first sound of the first test question from the broadcasting module, and the server randomly selects a first image from the first image library corresponding to the first sound, and then selects the first sound The picture is displayed on the display module at the same time, and then the subject selects a first picture corresponding to the first sound according to the first sound, and then generates a correspondence according to the first picture selected by the subject Enter the content of the first answer to the first test question and enter the processing module.

The calculation step includes a first calculation sub-step. In the first calculation sub-step, the processing module generates a first result based on a ratio of a first standard answer corresponding to the first test question to the first answer And stored in the storage module.

In addition, another object of the present invention is to provide an automated auditory perception evaluation device for performing the aforementioned automated auditory perception evaluation method.

The effect of the present invention is to use the evaluation device to select and execute the first test question for the subject to select a first picture according to the first sound, and the first answer content and the first result can be generated in sequence , And the first result is used to evaluate the subject's ability to perceive concepts. Because the assessment is only based on hearing and vision, it does not need to be applied to the subject's oral ability to perform the assessment. Therefore, for patients who cannot be expressed in spoken language, more accurate assessment results can be produced.

Referring to FIGS. 1 and 2, an embodiment of the automated hearing perception evaluation method of the present invention is suitable for providing a subject to be tested, and includes a preparation step 11, an evaluation step 12, and a calculation step 13. The preparation step 11 provides an evaluation device 2 that includes a server 21 and a user terminal 22 that connects to the server 21. The server 21 includes a data module 211 storing multiple first questions and multiple second questions, a processing module 212 with information connected to the data module 211, and a storage with information connected to the processing module 212 Module 213. The user terminal 22 includes a broadcast module 221 with information connected to the data module 211, a display module 222 with information connected to the data module 211 and the processing module 212, and a display module 222 connected to the information with the information The input module 223 of the processing module 212.

The first test question of each question has a first sound, and a plurality of first image libraries that are related to the first sound and each have a plurality of first pictures. In this embodiment, the first sounds are environmental sounds that express a message by non-voices emitted by humans or non-humans, such as non-verbal sounds produced by human activities, natural sounds, tool sounds, or environmental background sounds, etc. .

The second test question of each question has a second sound, and multiple questions correspond to the second sound, and there are second sub-test questions of multiple second libraries respectively. In this embodiment, the second test question of each question has four questions. Two sub-questions. Among them, the types of the questions of the second sub-examination questions are places, characters, objects, action situations or associative things, and each second gallery has multiple second pictures. In this embodiment, the second sounds express a situation with at least one non-speech sound produced by humans or non-humans, or express at least one non-speech sound produced by humans or non-humans in conjunction with at least one speech Situational sounds of a situation, such as the sounds around when shopping or riding a car.

In this embodiment, the first test questions of each question respectively have four first picture galleries, and the second sub-test questions of the type of the place, person, item, or action situation have four second picture galleries, each of the first picture galleries. And each second image library represents the content corresponding to the first sound or the second sound, semantically related to the first sound or the second sound content, and acoustic characteristics related to the first sound or the second sound, and There are four different meanings that are not related to the first sound or the second sound content.

The second sub-questions with a question type of associative things have two second galleries, one of which is related to the second sound content of the second examination question, and the other second gallery The second pictures are not related to the second sound content of the second test question. However, in actual implementation, the number and representative meanings of the first and second libraries can be adjusted according to the evaluation needs, not limited to the above-mentioned forms. Each gallery has multiple pictures related to the meaning of the representative, so as to avoid the subject from remembering the correct answer in the evaluation process of this embodiment multiple times, thereby affecting the evaluation results and improving the evaluation of this embodiment Correctness.

In the first test question of each question and the second sub-test question of each question, the first gallery matching the first sound content and the second gallery matching the second sound content respectively have the first sound Or the object picture of the second sound, for example, when the first sound or the second sound is the sound of an ambulance, there is a picture of the ambulance in the corresponding first gallery or the second gallery. The first image library semantically related to the first sound content and the second image library semantically related to the second sound content are object images with potential semantic meaning related to the ambulance sound, such as first aid box, syringe, etc. .

In the first image library related to the acoustic characteristics of the first sound and the second image library related to the acoustic characteristics of the second sound, the acoustic characteristics such as the frequency and duration of the ambulance sound are adjusted with acoustic synthesis software, and then Adjusted sound The ambulance sound finds another sound similar to it, and puts the picture of the object that emits the sound, such as a police car, fire truck, etc. In the first gallery that is not related to the first sound content and the second gallery that is not related to the second sound content, there is the exclusion of the corresponding first sound or the corresponding second sound content Images that are consistent, semantically related, and acoustically related, such as computers, televisions, etc. In the second gallery related to the content of the second sound, there are object pictures that can be associated when hearing the second sound, such as physicians, nurses, etc.

In the evaluation step 12, in order to ensure that the subject has fully understood the evaluation process, the evaluation process and questions can be displayed in text on the use end 22 of the evaluation device 2 before the subject is formally tested. And play pre-recorded guidance voice assistant instructions. Then, the server 21 demonstrates a demonstration question, and then the subject answers an exercise question to confirm the response of the subject before the subject is formally tested.

In order to make the evaluation results more comprehensive and more discriminative, in the evaluation step 12, different subject types will be used to evaluate the subject's basic and advanced perception concepts. Therefore, the evaluation step 12 includes a first quantum evaluation step 121 and a second quantum evaluation step 122, and in the first quantum evaluation step 121 and the second quantum evaluation step 122, different types of The first test question and the second test question are evaluated.

In the first quantum evaluation step 121, the server 21 randomly selects a first test question from the data module 211 and transmits it to the user terminal 22 to be answered by the subject. The user terminal 22 plays the first sound of the first test question from the broadcast module 221, and the server 21 randomly selects a first picture from the first image library corresponding to the first sound, and then selects When the first pictures are displayed on the display module 222 at the same time, the display positions of the first pictures are randomly distributed to prevent the subject from responding according to the regularity of the display, thereby affecting the accuracy of the evaluation results. Then, after the first sound is played, the subject selects a picture of the object that emits the first sound from the first pictures using the input module 223 within a limited time, and then according to the subject The selected first picture generates a first answer content corresponding to the first test question, and the first answer content is input to the processing module 212.

In the second quantum evaluation step 122, the server 21 randomly selects a second test question from the data module 211 and transmits it to the user terminal 22 for the subject to answer. The user terminal 22 is selected from the broadcast mode Group 221 plays the second sound of the second test question, and the server 21 randomly selects a second picture in the second image library corresponding to one of the second sub-test questions of the second sound, and then selects When the second pictures are displayed on the display module 222 at the same time, the display positions of the second pictures are randomly distributed to prevent the subject from responding according to the regularity of the display, thereby affecting the accuracy of the evaluation results.

After the second sound is played, the subject selects one of the second pictures directly related to the second sound using the input module 223 according to the question type of the second sub-test within the limited time Or pictures of unrelated objects. If the topic type of the second sub-test is a place, person, object, or action situation, the subject must select an object picture directly related to the second sound, and if the topic type is an associative thing, the subject must Select an object picture that is not related to the second sound. According to the second picture selected by the subject, a second answer content corresponding to the second sub-test is generated, and the second answer content is input to the processing module 212.

When performing the evaluation step 12, if the subject does not answer the question within the limited time after the first sound or the second sound is played, the subject requests to play the first sound or the second sound again, Or when playing the first sound or the second sound is interfered by external noise or things, the subject may request to play the first sound or the second sound again and use the subject to answer again The selected first picture or the second picture is the content of the answer, so as to exclude external factors that may affect the evaluation result and improve the accuracy of the evaluation result. In the evaluation step 12, the first evaluation quantum step 121 and the second evaluation quantum step 122 may be performed multiple times to allow the subject to answer multiple first test questions and multiple second test questions and generate multiple One answer content and multiple second answer content, by increasing the number of the first answer content and the second answer content that can be analyzed, the accuracy of the evaluation result is improved. During the actual test, the number of the first test questions and the second test questions to be tested can be adjusted according to the subject's concentration ability, so as to avoid affecting the assessment result due to the subject's insufficient attention when answering.

The calculation step 13 includes a first calculation sub-step 131 and a second calculation sub-step 132. In the first calculation sub-step 131, the processing module 212 compares the first answer content with a first standard answer corresponding to the first test question, and generates a representative meaning that is consistent with the first sound content, The first result that is semantically related to the first sound content, related to the acoustic characteristics of the first sound, or not related to the first sound content is stored in the storage module 213. In the second calculation sub-step 132, the processing module 212 generates a representative meaning that is consistent with the second sound content based on a second standard answer ratio corresponding to the second sub-test question to the second answer content, The second result that is semantically related to the second sound content, related to the acoustic characteristics of the second sound, not related to the second sound content, or related to the second sound content is stored in the storage module 213.

If the first result is consistent with the first sound content, or the topic type is location, person, object or action situation, the second result is consistent with the second sound content, it means that the subject has selected Positive answers to questions. Or the second result that the type of the topic is associative things is not related to the second sound content, which means that the subject’s cognition can judge all the options and choose the only one that has nothing to do with the second sound content. The negative answer can still be used to judge that the subject has the correct perception concept. The three situations mentioned above can be regarded as one correct answer, and each correct answer can get one point. The higher the total score obtained by the subject, the more complete the perception concept is.

By performing the above steps with the evaluation device 2, the evaluation process can be standardized, and the subject’s perceptual conceptual ability can be quantified, making the evaluation results more objective, and a tester does not need to spend a lot of time learning the evaluation process and scoring method That is, it can be tested in this embodiment, and the clinical practicality and efficiency of this embodiment can be implemented.

Referring to FIG. 3, it is one of the execution modes of this embodiment, taking the first sound as an ambulance sound as an example. The subject hears the sound of the ambulance played by the first test question, and can simultaneously see a picture 31 of an ambulance in the first gallery that matches the content of the first sound, and the semantically related to the content of the first sound A picture 32 of a first aid kit in the first gallery, a picture 33 of a police car in the first gallery related to the acoustic characteristics of the first sound, and one of the first gallery not related to the content of the first sound Picture 34 of the computer. The subject must then select a picture of the object that emits the sound of an ambulance from these pictures 31 to 34.

Referring to FIG. 4, taking the second sound as an aircraft landing sound as an example. When the subject heard the landing sound of the second test question and answered the second sub-test question with the question type as location, he could also see a picture of an airport in the second gallery matching the content of the second sound 41. A picture of a passenger station in the second gallery semantically related to the second sound content 42, a picture of a garage in the second gallery related to the acoustic characteristics of the second sound 43, and a picture related to the second Picture 44 of a classroom in the second gallery that is not related to the content of the second sound. The subject must then select a picture of the place where the landing sound of the aircraft can be heard from these pictures 41-44.

Referring to FIG. 5, when the subject answers the second sub-question with a question type of associative thing, he can see a picture 51 of an airplane in the second gallery related to the second sound content, and a luggage Picture 52, picture 53 of a flight ticket, and picture 54 of a cow in the second gallery not related to the second sound content. The subject must then choose from these pictures 51 to 54 a picture of something that would not be associated with the sound of the plane landing.

Referring back to FIGS. 3 to 5, in the aforementioned first test question, if the corresponding first response content is the picture 31 of the ambulance, the first result generated is the first sound content meets the. In the second sub-question with the aforementioned question type as location, if the corresponding second answer content is the picture 41 of the airport, the second result generated is consistent with the second sound content. In the second sub-question where the aforementioned question type is an associative thing, if the corresponding second answer content is the picture 54 of the cow, the second result generated is not related to the second sound content . In the aforementioned three situations, the subject can get one point each, indicating that the subject's cognition of the first sound or the second sound is correct.

If the first or the second answers are other pictures, it means that the subject’s cognition of the first sound or the second sound will be based on the first gallery and the corresponding ones of the first pictures The representative meaning of the second image library corresponding to the second picture, and there are different types of errors, the tester can further determine the cognitive status of the subject according to the distribution of different representative meanings. For example, the first result and the second results are concentrated on the acoustic characteristics of the first sound or the second sound, which may indicate that the auditory processing function of the subject has become abnormal and the correct sound signal cannot be transmitted High-level central auditory processing center, which in turn affects answering. If the first result and the second results are dispersed into different representative meanings and cannot display a specific trend, it indicates that the subject may answer more by guessing due to factors such as inattention. If the first result is consistent with the first sound content, but the second results are different types of errors, it means that the subject may still have basic perceptual conceptual abilities, but high-level perceptual conceptual abilities already exist Part of the damage, which in turn affects the subject's auditory perception performance.

In summary, the present invention uses the evaluation device 2 to select and execute the test questions for the subject to select a picture according to the first sound or the second sound played, and then a response content can be generated in sequence And a result, and use the result to evaluate the subject's perception of conceptual ability. During the process, only the auditory and visual assessment is used, and the subject’s oral ability can be used for assessment. For patients who cannot express in oral language, a more accurate assessment result can be produced, so it can indeed achieve cost invention. Purpose.

However, the above are only examples of the present invention, and should not be used to limit the scope of the present invention. Any simple equivalent changes and modifications made according to the scope of the patent application of the present invention and the content of the patent specification are still classified as This invention covers the patent.

11‧‧‧Preparation step 12‧‧‧Evaluation step 121‧‧‧ First evaluation quantum step 122‧‧‧ Second evaluation quantum step 13‧‧‧Calculation step 131‧‧‧First calculation substep 132‧‧‧ Second calculation substep 2‧‧‧Evaluation device 21‧‧‧Servo terminal 211‧‧‧Data module 212‧‧‧Processing module 213‧‧‧Storage module 22‧‧‧Using terminal 221‧‧‧ Broadcasting module Group 222‧‧‧ Display module 223‧‧‧ Input module 31~34‧‧‧Picture 41~44‧‧‧Picture 51~54‧‧‧Picture

Other features and functions of the present invention will be clearly presented in the embodiments with reference to the drawings, in which: FIG. 1 is a flowchart illustrating an embodiment of the automated auditory perception evaluation method of the present invention; FIG. 2 is a block function Fig. 3 illustrates an evaluation device provided in a preparatory step of this embodiment; Fig. 3 is a schematic diagram illustrating the execution method of a first quantum evaluation step of this embodiment; Fig. 4 is a schematic diagram illustrating the embodiment of this embodiment A second quantum review step, and explain the execution method of one of the second sub-test questions; and FIG. 5 is a schematic diagram illustrating the execution method of another second sub-test question in the second quantum step.

11‧‧‧Preparation steps

12‧‧‧Evaluation steps

121‧‧‧ First review of quantum steps

122‧‧‧ Second review of quantum steps

13‧‧‧Calculation steps

131‧‧‧ First calculation substep

132‧‧‧Second calculation substep

Claims (10)

An automated auditory perception evaluation method, suitable for providing a subject to be tested, and includes: a preparation step, providing an evaluation device, the evaluation device including a server, and a user terminal for information connection to the server, the server It includes a data module storing plural first test questions, a processing module with information connected to the data module, and a storage module with information connected to the processing module. The user terminal includes an information connected to the data module Broadcasting module, an information module connecting the data module and the processing module, and an information module connecting the display module and the processing module, each first question has a first sound , And a first gallery whose plural numbers are related to the first sound and have plural first pictures, respectively; an evaluation step, including a first quantum evaluation step, in which the servo terminal selects from the data model Select a first test question in the group and transmit it to the user end to be answered by the subject. The user end plays the first sound of the first test question from the broadcast module, and the server end corresponds to the first A first picture is selected from the first image libraries of the sound, and the first pictures are displayed on the display module at the same time, and then the subject selects correspondingly through the input module according to the first sound A first picture, and then generate a first answer corresponding to the first test question based on the first picture selected by the subject, and input the processing module; and a calculation step, including a first calculator Step, in the first calculation sub-step, the processing module generates a first result according to a first standard answer ratio corresponding to the first test question to the first answer content, and stores it in the storage module . The automated auditory perception evaluation method according to claim 1, wherein the data module further stores a plurality of second test questions, each of the second test questions has a second sound, and the plural corresponds to the second sound and has a plural number respectively The second sub-test questions of the second gallery, each second gallery has a plurality of second pictures, the assessment step also includes a second assessment quantum step, in the second assessment quantum step, the server is from the data module Select a second test question and transmit it to the user end to be answered by the subject. The user end plays the second sound of the second test question from the broadcast module, and the server end responds to the second sound In the second image gallery of a second sub-test, select a second picture respectively, and then display the second pictures on the display module at the same time, according to the second sound The input module selects a pair of second pictures corresponding to the second sub-test question, then generates a second answer content corresponding to the second sub-test question based on the second picture selected by the subject, and enters the processing module The calculation step also includes a second calculation sub-step. In the second calculation sub-step, the processing module generates an answer based on a second standard answer ratio corresponding to the second sub-test question to the second answer The second result is stored in the storage module. The automated auditory perception evaluation method according to claim 1 or 2, wherein the first sounds are non-speech sounds emitted by humans or non-humans, and express environmental sounds of a message. The automated auditory perception evaluation method according to claim 1 or 2, wherein the first sounds are selected from non-speech sounds of human beings, natural sounds, tool sounds, environmental background sounds, or a combination thereof. The automated auditory perception evaluation method according to claim 2, wherein the second sounds are situational sounds expressing a situation with at least one non-speech sound emitted by a human or a non-human. The automated auditory perception evaluation method according to claim 2, wherein the second sounds are situational sounds expressing a situation in conjunction with at least one non-speech sound emitted by a human or non-human being and at least one speech sound. The automated auditory perception evaluation method according to claim 1 or 2, wherein in the evaluation step, the subject must select a corresponding first picture within a limited time after the first sound is played . The automated auditory perception evaluation method according to claim 2, wherein in the evaluation step, after the second sound is played, the subject must select a corresponding second picture within the limited time. The automated auditory perception evaluation method according to claim 2, wherein the types of questions of the second sub-test questions are selected from places, characters, objects, action situations, associative things, or a combination thereof. An automated auditory perception evaluation device for performing the automated auditory perception evaluation method according to claim 1.

Images