Spoken dialogue quality evaluation method for college English learning

Abstract

The invention discloses a spoken language dialogue quality evaluation method for college English learning, and relates to the technical field of data processing, and the method comprises the steps: constructing a dialogue knowledge database; dividing the spoken dialogue voice into actual dialogue boxes; identifying to obtain a dialogue application scene; obtaining a semantic misunderstanding part; obtaining a misunderstanding degree coefficient; analyzing to obtain the pronunciation problem degree and the reaction initiating speed of the actual dialog box; comprehensively obtaining a dialogue continuity index; obtaining a quality evaluation value; obtaining a quality evaluation value range by using the sample dialogue content; calculating the proportion of the part, exceeding the quality evaluation value range, of the quality evaluation value as an identification proportion; and when the identification proportion is larger, the spoken language dialogue quality is lower. A dialogue application scene is obtained through recognition, a semantic misunderstanding part is obtained, a misunderstanding degree coefficient is obtained, and the pronunciation problem degree and the reaction initiating speed are analyzed, so that spoken dialogues can be evaluated more accurately from multiple aspects.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and particularly to a method for evaluating the quality of oral conversations for college English learning. Background Art

[0002] With the development of globalization and the deepening of exchanges and communications among countries, multilingual learning has been increasingly emphasized, and mastering a foreign language is very important. Among them, English, as a widely used foreign language, is included in the basic courses of higher education in China. In the process of English teaching in colleges and universities, due to problems such as a large number of students, weak interactivity, a single teaching mode, and language environment, an imbalance problem between the written test education part and the oral education part has occurred in the English education process. Students often have high written test scores but lack the oral English ability to communicate in English.

[0003] Due to the variability of oral conversations, it is difficult to accurately evaluate the quality of oral conversations, resulting in the inability to conduct targeted training for students. Summary of the Invention

[0004] To solve the above technical problems, a method for evaluating the quality of oral conversations for college English learning is provided, and this technical solution solves the problems raised in the above background art.

[0005] To achieve the above object, the technical solution adopted by the present invention is as follows:

[0006] A method for evaluating the quality of oral conversations for college English learning, including:

[0007] Based on the content of at least one sample conversation of native English speakers, a conversation knowledge database is constructed;

[0008] The oral conversation voice is obtained, and according to the time sequence, the oral conversation voice is segmented into at least one actual dialogue box;

[0009] Based on the oral conversation voice and the conversation knowledge database, the conversation application scenario is identified;

[0010] Based on the conversation application scenario, the identification of unreasonable content is carried out to obtain the semantic misunderstanding part;

[0011] The unreasonable degree of the semantic misunderstanding part is analyzed to obtain the misunderstanding degree coefficient;

[0012] The pronunciation problem degree and the initiation response speed of the actual dialogue box are analyzed;

[0013] Based on the analysis of the proportion of semantic misunderstandings, the conversation coherence index is comprehensively obtained;

[0014] Form the weights of the pronunciation problem degree, the initiation reaction speed, and the dialogue coherence index, and use the corresponding weights to superimpose the pronunciation problem degree, the initiation reaction speed, and the dialogue coherence index to obtain a quality evaluation value;

[0015] Use the sample dialogue content to obtain the range of the quality evaluation value;

[0016] Calculate the proportion of the part where the quality evaluation value exceeds the range of the quality evaluation value as the recognition ratio;

[0017] When the recognition ratio is larger, the quality of the spoken dialogue is lower.

[0018] Preferably, constructing a dialogue knowledge database based on at least one sample dialogue content of native English speakers includes the following steps:

[0019] Based on the voice timbre, segment the sample dialogue content in chronological order to obtain at least one sample dialogue box. The sample dialogue box contains the spoken content of the same person, and the spoken content contained in adjacent sample dialogue boxes belongs to different people;

[0020] Sort the sample dialogue boxes of the sample dialogue content in chronological order to obtain a sample dialogue sequence, and number the sample dialogue boxes in the sample dialogue sequence in order;

[0021] Use a neural network model to construct a semantic recognition model;

[0022] Use the semantic recognition model to recognize the semantics of the sample dialogue boxes in the sample dialogue sequence to obtain the sample dialogue box semantics;

[0023] When there is an association between the sample dialogue box semantics of two sample dialogue boxes, establish a corresponding relationship between the two sample dialogue boxes;

[0024] Form at least one associated dialogue set. The associated dialogue set is composed of sample dialogue boxes of the same sample dialogue content, and any two sample dialogue boxes in the associated dialogue set are in a corresponding relationship;

[0025] Perform semantic recognition on the sample dialogue content that generates the associated dialogue set to obtain the sample semantics;

[0026] In the big data, retrieve at least one dialogue text that expresses the same semantics as the sample semantics, obtain the occurrence context of the dialogue text, and extract the features of the occurrence context to obtain at least one application feature;

[0027] Count the number of occurrences of the application feature as the weight of the application feature;

[0028] Accumulate the weights of the application features of the occurrence context to obtain a judgment value;

[0029] Take the occurrence scenario with the largest judgment value as the occurrence scenario of the associated dialogue set, and summarize the associated dialogue set and its corresponding occurrence scenario into a dialogue knowledge database.

[0030] Preferably, the step of segmenting the spoken dialogue speech into at least one actual dialogue box according to the time sequence includes the following steps:

[0031] Based on the timbre, segment the spoken dialogue speech according to the time sequence to obtain at least one actual dialogue box. The actual dialogue box contains the spoken content of the same person, and the spoken content contained in adjacent actual dialogue boxes belongs to different people. Number the actual dialogue boxes in chronological order.

[0032] Preferably, the steps of identifying the dialogue application scenario based on the spoken dialogue speech and the dialogue knowledge database include the following steps:

[0033] Perform semantic recognition on the actual dialogue boxes of the spoken dialogue speech to obtain the actual dialogue box semantics;

[0034] When there is an association between the actual dialogue box semantics of two actual dialogue boxes, establish a corresponding relationship between the two actual dialogue boxes;

[0035] Form at least one actual dialogue set. The actual dialogue set is composed of actual dialogue boxes, and any two actual dialogue boxes in the actual dialogue set are in a corresponding relationship;

[0036] Obtain the associated dialogue set with the largest overlapping part with the actual dialogue set, match it to the actual dialogue set, and match the occurrence scenario of the associated dialogue set to the corresponding actual dialogue set;

[0037] Deduplicate and summarize the occurrence scenarios corresponding to the actual dialogue sets of the spoken dialogue speech to obtain the dialogue application scenario.

[0038] Preferably, the steps of identifying unreasonable content to obtain the semantic misunderstanding part include the following steps:

[0039] Take the associated dialogue set corresponding to the occurrence scenario with the largest overlap with the dialogue application scenario as the preliminary associated dialogue set;

[0040] Take one of at least one actual dialogue box as the target actual dialogue box;

[0041] Take the actual dialogue set containing the target actual dialogue box as the target actual dialogue set;

[0042] Take the preliminary associated dialogue set with the same number of elements as the target actual dialogue set as the secondary associated dialogue set;

[0043] Take the secondary associated dialogue set with the largest overlapping part with the target actual dialogue set as the target associated dialogue set;

[0044] Arrange the sample dialog boxes in the target associated dialog set in sequence to obtain a first sequence, and arrange the actual dialog boxes in the target actual dialog set in sequence to obtain a second sequence;

[0045] Establish a corresponding relationship between the sample dialog box and the actual dialog box at the same position in the first sequence and the second sequence, and use the sample dialog box corresponding to the target actual dialog box as the target sample dialog box;

[0046] Obtain the part that is different from the target sample dialog box in the target actual dialog box as the suspicious part;

[0047] Based on big data, obtain the commonly used keywords in spoken language, classify the keywords, and the semantics of the keywords in the same category are the same;

[0048] Replace the suspicious part with keywords in the same category to obtain all replacement situations after the replacement of the suspicious part;

[0049] In each replacement situation, re-obtain the part that is different from the target sample dialog box in the target actual dialog box as the unreasonable part, and re-obtain the part that is the same as the target sample dialog box in the target actual dialog box as the reasonable part;

[0050] Take the unreasonable part with the smallest length as the semantic misunderstanding part;

[0051] When the target actual dialog box traverses at least one actual dialog box, at least one semantic misunderstanding part is obtained.

[0052] Preferably, the analysis of the unreasonable degree of the semantic misunderstanding part to obtain the misunderstanding degree coefficient includes the following steps:

[0053] Remove the reasonable part corresponding to the semantic misunderstanding part in the target sample dialog box to obtain a reference part;

[0054] Based on big data, count the probability of two keywords appearing synchronously as the correlation degree of the two keywords;

[0055] Obtain all possible combinations of the keywords in the semantic misunderstanding part and the keywords in the reference part to obtain at least one keyword group;

[0056] Take the reciprocal of the average value of the correlation degrees of the keywords in at least one keyword group to obtain the misunderstanding degree coefficient.

[0057] Preferably, the analysis to obtain the pronunciation problem degree and the response initiation speed of the actual dialog box includes the following steps:

[0058] Compare the standard pronunciation of English words with the sample dialog content to obtain at least one sample pronunciation;

[0059] Statistically determine the minimum value of the time intervals between adjacent sample pronunciations in chronological order as the characteristic time;

[0060] Perform pronunciation interval recognition on the content in the actual dialog box, and use the points where the pronunciation interval time is greater than the characteristic time as the segmentation points;

[0061] Take the content in the actual dialog box between adjacent segmentation points as the actual word pronunciation;

[0062] Obtain the words whose pronunciation difference from the actual word pronunciation is less than a preset value as the suspected words, summarize the set of suspected words, and use the suspected word with the smallest pronunciation difference from the actual word pronunciation in the set of suspected words as the target word. Replace the position of the actual word pronunciation in the actual dialog box with the target word to obtain the actual modified dialog box. The preset value is a parameter set based on empirical values;

[0063] When the semantics of the actual modified dialog box is consistent with that of the actual dialog box, then statistically determine the difference between the standard pronunciation of the target word and the actual word pronunciation as the degree of local problem. Otherwise, delete the target word from the set of suspected words and repeat the previous step until the degree of local problem is statistically determined;

[0064] Accumulate the degrees of local problems to obtain the degree of pronunciation problem;

[0065] Take the average value of the time intervals between adjacent actual dialog boxes to obtain the initiation response speed.

[0066] Preferably, the analysis based on the proportion of semantic misunderstandings to comprehensively obtain the dialogue coherence index includes the following steps:

[0067] Multiply the proportion of the semantically misunderstood part in the spoken dialogue speech by the misunderstanding degree coefficient of the semantically misunderstood part and accumulate to obtain the dialogue coherence index.

[0068] Preferably, the formation of the weights of the degree of pronunciation problem, initiation response speed, and dialogue coherence index includes the following steps:

[0069] Based on the analytic hierarchy process, form the weights of the degree of pronunciation problem, initiation response speed, and dialogue coherence index.

[0070] Preferably, the obtaining of the quality assessment value range using the sample dialogue content includes the following steps:

[0071] Obtain the quality assessment value of the sample dialogue content in the same way as the quality assessment value of the spoken dialogue speech;

[0072] Use the maximum and minimum values of the quality assessment value of the sample dialogue content as endpoints to form the quality assessment value range.

[0073] Compared with the prior art, the present invention has the following beneficial effects:

[0074] By identifying the dialogue application scenario, the semantic misunderstanding part, the misunderstanding degree coefficient, and analyzing the pronunciation problem degree and the reaction speed, the important situations in the oral dialogue can be comprehensively considered, and the errors in semantic understanding and pronunciation problems can be focused on. Then, the oral dialogue can be evaluated more accurately from multiple aspects, and targeted training can be provided to students in a timely manner according to the situation of the dialogue. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 A schematic diagram of a flow chart of a method for evaluating oral dialogue quality for college English learning according to the present invention;

[0076] Figure 2 A schematic diagram of a process of constructing a conversation knowledge database based on at least one sample conversation content of a native English speaker according to the present invention;

[0077] Figure 3 A schematic diagram of a process of identifying a dialogue application scenario based on spoken dialogue speech and a dialogue knowledge database according to the present invention;

[0078] Figure 4 A schematic diagram of a flow chart of the present invention for identifying unreasonable content and obtaining a semantic misunderstanding portion;

[0079] Figure 5 A flow chart of analyzing the unreasonable degree of the semantic misunderstanding part and obtaining the misunderstanding degree coefficient of the present invention;

[0080] Figure 6 A flow chart of the present invention for analyzing and obtaining the degree of pronunciation problems and the speed of initiating response of an actual dialog box;

[0081] Figure 7 It is a flow chart of obtaining a quality assessment value range using sample conversation content according to the present invention. DETAILED DESCRIPTION

[0082] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0083] Reference Figure 1 As shown, a method for evaluating the quality of oral dialogue for college English learning includes:

[0084] Constructing a conversation knowledge database based on at least one sample conversation content of a native English speaker;

[0085] Obtain the spoken dialogue voice and split the spoken dialogue voice into at least one actual dialogue box in chronological order;

[0086] Based on the spoken dialogue voice and the dialogue knowledge database, identify the dialogue application scenario;

[0087] Based on the dialogue application scenario, identify unreasonable content to obtain the semantic misunderstanding part;

[0088] Analyze the unreasonable degree of the semantic misunderstanding part to obtain the misunderstanding degree coefficient;

[0089] Analyze and obtain the pronunciation problem degree and the initiation response speed of the actual dialogue box;

[0090] Based on the analysis of the proportion of semantic misunderstandings, comprehensively obtain the dialogue coherence index;

[0091] Form the weights of the pronunciation problem degree, the initiation response speed, and the dialogue coherence index, and use the corresponding weights to superimpose the pronunciation problem degree, the initiation response speed, and the dialogue coherence index to obtain the quality evaluation value;

[0092] Use the sample dialogue content to obtain the quality evaluation value range;

[0093] Calculate the proportion of the part where the quality evaluation value exceeds the quality evaluation value range as the identification proportion;

[0094] When the identification proportion is larger, the quality of the spoken dialogue is lower.

[0095] The most important part of the spoken language quality evaluation is the evaluation of the rationality of the dialogue semantic understanding. In some cases, although there are no grammar errors and the dialogue content has no obvious problems, but it is completely inconsistent with the current scenario, then it actually misses the point. Therefore, it is necessary to identify the degree of this situation. Secondly, it is necessary to identify the pronunciation problems. Since it is not clear which word is being said, there will be a greater obstacle to the identification of pronunciation problems. Therefore, corresponding steps need to be set to solve this problem. The reaction speed actually reflects the student's mastery of the language, so it also needs to be taken into consideration, but it is relatively easy to identify.

[0096] Refer to Figure 2 As shown, based on at least one sample dialogue content of native English speakers, the steps to construct the dialogue knowledge database are as follows:

[0097] Based on the timbre, split the sample dialogue content in chronological order to obtain at least one sample dialogue box. The sample dialogue box contains the spoken content of the same person, and the spoken content contained in adjacent sample dialogue boxes belongs to different people;

[0098] Sort the sample dialog boxes of the sample conversation content in chronological order to obtain a sample conversation sequence, and number the sample dialog boxes in the sample conversation sequence in order;

[0099] Use a neural network model to construct a semantic recognition model;

[0100] Use the semantic recognition model to recognize the semantics of the sample dialog boxes in the sample conversation sequence to obtain the sample dialog box semantics;

[0101] When there is an association between the sample dialog box semantics of two sample dialog boxes, establish a corresponding relationship between the two sample dialog boxes;

[0102] Form at least one associated conversation set, which is composed of sample dialog boxes of the same sample conversation content, and any two sample dialog boxes in the associated conversation set are in a corresponding relationship;

[0103] Perform semantic recognition on the sample conversation content that generates the associated conversation set to obtain the sample semantics;

[0104] In the big data, retrieve at least one conversation text with the same semantics as the sample semantics, obtain the occurrence context of the conversation text, and extract features from the occurrence context to obtain at least one application feature;

[0105] Count the number of occurrences of the application feature as the weight of the application feature;

[0106] Accumulate the weights of the application features of the occurrence context to obtain a judgment value;

[0107] Take the occurrence context with the largest judgment value as the occurrence context of the associated conversation set, and summarize the associated conversation set and its corresponding occurrence context into a conversation knowledge database.

[0108] The conversation knowledge database mainly forms its corresponding occurrence context for the associated conversation set. Subsequently, the situation can be judged according to the occurrence context, and then the associated conversation set with the same situation can be called. The associated conversation set is similar to a context relationship because each conversation sentence may not only be a reply to the previous sentence but also a reply to the previous few sentences. Therefore, it is necessary to identify the associated conversations and consider them together.

[0109] Segment the spoken dialogue speech into at least one actual dialog box in chronological order, including the following steps:

[0110] Based on the timbre, segment the spoken dialogue speech in chronological order to obtain at least one actual dialog box. The actual dialog box contains the spoken content of the same person, and the spoken content contained in adjacent actual dialog boxes belongs to different people. Number the actual dialog boxes in chronological order.

[0111] Spoken dialogue voice is carried out by several people, and each person takes turns to communicate in the dialogue. Therefore, the dialogue is segmented according to the timbre, and through the acquisition of the actual dialogue box, subsequent analysis can be facilitated.

[0112] Refer to Figure 3 As shown, based on the spoken dialogue voice and the dialogue knowledge database, the steps for identifying the dialogue application scenario include the following:

[0113] Perform semantic recognition on the actual dialogue box of the spoken dialogue voice to obtain the actual dialogue box semantics;

[0114] When there is an association between the actual dialogue box semantics of two actual dialogue boxes, a corresponding relationship is established between the two actual dialogue boxes;

[0115] Form at least one actual dialogue set, which is composed of actual dialogue boxes, and any two actual dialogue boxes in the actual dialogue set are in a corresponding relationship;

[0116] Obtain the associated dialogue set with the largest overlapping part with the actual dialogue set, match it to the actual dialogue set, and match the occurrence scenario of the associated dialogue set to the corresponding actual dialogue set;

[0117] Deduplicate and summarize the occurrence scenarios corresponding to the actual dialogue set of the spoken dialogue voice to obtain the dialogue application scenario.

[0118] When performing semantic analysis, first of all, it is necessary to determine the scenario of the dialogue. Otherwise, without the limitation of the scenario, it is very easy to analyze wrongly. Because, in different scenarios, different dialogues may be reasonable or unreasonable. Therefore, when the scenario cannot be determined, it is not convenient to analyze the rationality of the dialogue.

[0119] Refer to Figure 4 As shown, the steps for identifying unreasonable content to obtain the semantic misunderstanding part include the following:

[0120] Take the associated dialogue set corresponding to the occurrence scenario with the largest overlap with the dialogue application scenario as the preliminary associated dialogue set;

[0121] Take one of at least one actual dialogue box as the target actual dialogue box;

[0122] Take the actual dialogue set containing the target actual dialogue box as the target actual dialogue set;

[0123] Take the preliminary associated dialogue set with the same number of elements as the target actual dialogue set as the secondary associated dialogue set;

[0124] Take the secondary associated dialogue set with the largest overlapping part with the target actual dialogue set as the target associated dialogue set;

[0125] Arrange the sample dialog boxes in the target associated dialogue set in sequence to obtain a first sequence, and arrange the actual dialog boxes in the target actual dialogue set in sequence to obtain a second sequence;

[0126] Establish a corresponding relationship between the sample dialog box and the actual dialog box at the same position in the first sequence and the second sequence, and use the sample dialog box corresponding to the target actual dialog box as the target sample dialog box;

[0127] Obtain the part that is different from the target sample dialog box in the target actual dialog box as the suspicious part;

[0128] Based on big data, obtain the commonly used keywords in spoken language, classify the keywords, and the semantics of the keywords in the same category are the same;

[0129] Replace the suspicious part with the keywords in the same category to obtain all replacement situations after the replacement of the suspicious part;

[0130] In each replacement situation, re-obtain the part that is different from the target sample dialog box in the target actual dialog box as the unreasonable part, and re-obtain the part that is the same as the target sample dialog box in the target actual dialog box as the reasonable part;

[0131] Take the unreasonable part with the minimum length as the semantic misunderstanding part;

[0132] When the target actual dialog box traverses at least one actual dialog box, at least one semantic misunderstanding part is obtained.

[0133] Since the situations of daily conversations are relatively limited, therefore, by constructing a dialogue knowledge database, all possibilities can be approximated. Although not all possibilities can be included, basically, actual conversations can find approximately corresponding situations in the dialogue knowledge database. Thus, the target associated dialogue set is obtained. The structure of the target associated dialogue set is the most similar to that of the target actual dialogue set, and the scenarios are also the closest. Since the elements in the target associated dialogue set and the target actual dialogue set are arranged in chronological order, therefore, the target sample dialog box with the same relative position can be found for comparison with the target actual dialog box, and then the unreasonable part can be obtained. However, due to the existence of synonyms, therefore, replacement comparison is required to finally determine the semantic misunderstanding part.

[0134] Refer to Figure 5 As shown, analyze the unreasonable degree of the semantic misunderstanding part, and the steps for obtaining the misunderstanding degree coefficient include:

[0135] Remove the reasonable part corresponding to the semantic misunderstanding part from the target sample dialog box to obtain the reference part;

[0136] Based on big data, the probability of two keywords appearing synchronously is statistically calculated as the correlation degree between the two keywords;

[0137] Obtain all possible combinations of the keywords in the semantic misunderstanding part and the keywords in the reference part to get at least one keyword group;

[0138] Take the reciprocal of the average value of the correlation degrees of the keywords in at least one keyword group to obtain the misunderstanding degree coefficient.

[0139] Taking the average value of the correlation degree is an evaluation of the correlation. Therefore, when its reciprocal is taken, it is an evaluation of the non - correlation. Therefore, it can be understood as the misunderstanding degree coefficient.

[0140] Refer to Figure 6 As shown, analyzing to obtain the pronunciation problem degree and the initiation response speed of the actual dialog box includes the following steps:

[0141] Compare the standard pronunciation of English words with the sample dialogue content to obtain at least one sample pronunciation;

[0142] Statistically calculate the minimum value of the time intervals between adjacent sample pronunciations in chronological order as the characteristic time;

[0143] Identify the pronunciation intervals in the content of the actual dialog box, and take the points where the pronunciation interval time is greater than the characteristic time as the segmentation points;

[0144] Take the content in the actual dialog box between adjacent segmentation points as the actual word pronunciation;

[0145] Obtain the words whose pronunciation gap from the actual word pronunciation is less than the preset value as the suspected words, summarize the set of suspected words, and take the suspected word with the smallest pronunciation gap from the actual word pronunciation in the set of suspected words as the target word. Use the target word to replace the position where the actual word pronunciation is in the actual dialog box to obtain the actual modified dialog box. The preset value is a parameter set based on empirical values;

[0146] When the semantics of the actual modified dialog box is consistent with that of the actual dialog box, then statistically calculate the gap between the standard pronunciation of the target word and the actual word pronunciation as the local problem degree. Otherwise, delete the target word from the set of suspected words and repeat the previous step until the local problem degree is statistically calculated;

[0147] Accumulate the local problem degrees to obtain the pronunciation problem degree;

[0148] Take the average value of the time intervals between adjacent actual dialog boxes to obtain the initiation response speed.

[0149] The preset value is set based on experience. When the pronunciation difference does not exceed the preset value, the two pronunciations can be considered to be the same pronunciation. This can be obtained based on experience testing;

[0150] Since the pronunciation of the actual word may not be standard, it is necessary to determine the actual corresponding word, and then compare the actual word pronunciation with the actual corresponding word to identify the pronunciation problem.

[0151] Based on the analysis of the proportion of semantic misunderstandings, the dialogue coherence index is obtained by the following steps:

[0152] The proportion of the semantic misunderstanding part in the spoken conversation speech is multiplied and added with the misunderstanding degree coefficient of the semantic misunderstanding part to obtain the conversation coherence index.

[0153] The weights for forming the pronunciation problem level, initiation response speed, and conversation coherence indexes include the following steps:

[0154] Based on the analytic hierarchy process, weights for the degree of pronunciation problems, speed of initiation response, and conversation coherence index were formed.

[0155] Reference Figure 7 As shown, obtaining the quality assessment value range using the sample conversation content includes the following steps:

[0156] Obtaining a quality evaluation value of the sample conversation content in the same manner as the quality evaluation value of the spoken conversation speech;

[0157] The maximum and minimum values ​​of the quality evaluation values ​​of the sample conversation contents are used as endpoints to form a quality evaluation value range.

[0158] Furthermore, the present solution also proposes a storage medium on which a computer-readable program is stored. When the computer-readable program is called, the above-mentioned oral dialogue quality assessment method for college English learning is executed.

[0159] It is understandable that the storage medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).

[0160] In summary, the advantages of the present invention are: by identifying the dialogue application scenario, obtaining the semantic misunderstanding part, obtaining the misunderstanding degree coefficient, and analyzing the pronunciation problem degree and the reaction speed, the important situations in the oral dialogue are comprehensively considered, and the errors in semantic understanding and pronunciation problems are focused on, so as to more accurately evaluate the oral dialogue from multiple aspects, and provide targeted training to students in a timely manner according to the situation of the dialogue.

[0161] The foregoing has shown and described the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and what is described in the above embodiments and the specification is only the principle of the present invention. Without departing from the spirit and scope of the present invention, various changes and improvements will occur to the present invention, and these changes and improvements fall within the scope of the present invention claimed. The scope of protection required by the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for evaluating the quality of oral dialogues in college English learning, characterized in that: include: Constructing a conversation knowledge database based on at least one sample conversation content of a native English speaker; Acquire spoken conversation speech, and segment the spoken conversation speech into at least one actual dialog box in chronological order; Based on the spoken conversation speech and conversation knowledge database, the conversation application scenario is identified; Based on the dialogue application scenario, the unreasonable content is identified to obtain the semantic misunderstanding part; Analyze the unreasonable degree of semantic misunderstanding and obtain the misunderstanding degree coefficient; Analyze the pronunciation problem level and the response speed of the actual dialog box; Based on the analysis of the proportion of semantic misunderstanding, the dialogue coherence index was obtained; forming weights of the pronunciation problem degree, the initiation reaction speed, and the conversation coherence index, and superimposing the pronunciation problem degree, the initiation reaction speed, and the conversation coherence index using the corresponding weights to obtain a quality assessment value; Use sample conversation content to obtain a quality assessment value range; Calculate the proportion of the parts whose quality assessment values ​​exceed the quality assessment value range as the recognition ratio; When the recognition ratio is larger, the quality of the spoken dialogue is lower.

2. A method for evaluating the quality of oral dialogue for college English learning according to claim 1, characterized in that: The step of constructing a conversation knowledge database based on at least one sample conversation content of a native English speaker comprises the following steps: Based on the timbre, the sample dialogue content is segmented in chronological order to obtain at least one sample dialog box, wherein the sample dialog box contains spoken content of the same person, and the spoken content contained in adjacent sample dialog boxes belongs to different people; Sorting the sample dialog boxes of the sample dialog contents in chronological order to obtain a sample dialog sequence, and numbering the sample dialog boxes in the sample dialog sequence in order; Use neural network model to build semantic recognition model; Using the semantic recognition model, the semantics of the sample dialog is recognized in the sample dialog sequence to obtain the semantics of the sample dialog; When the sample dialog box semantics of two sample dialog boxes are related, a corresponding relationship is established between the two sample dialog boxes; Forming at least one associated dialog set, the associated dialog set consisting of sample dialogs with the same sample dialog content, and any two sample dialogs in the associated dialog set are in a corresponding relationship; Perform semantic recognition on the sample conversation content that generates the associated conversation set to obtain sample semantics; In the big data, at least one dialogue text having the same semantics as the sample semantic expression is retrieved, the occurrence context of the dialogue text is obtained, and features of the occurrence context are extracted to obtain at least one application feature; Count the number of occurrences of application features as the weight of the application features; The weights of the application features of the occurrence situation are accumulated to obtain the judgment value; The occurrence scenario with the largest judgment value is taken as the occurrence scenario of the associated dialogue set, and the associated dialogue set and its corresponding occurrence scenarios are summarized into a dialogue knowledge database.

3. The method for evaluating the quality of oral dialogue for college English learning according to claim 2, characterized in that: The method of segmenting the spoken dialogue speech into at least one actual dialogue box in chronological order comprises the following steps: Based on the timbre, the spoken dialogue speech is segmented in chronological order to obtain at least one actual dialog box, wherein the actual dialog box contains spoken content of the same person, and the spoken content contained in adjacent actual dialog boxes belongs to different people, and the actual dialog boxes are numbered in chronological order.

4. A method for evaluating the quality of oral dialogue for college English learning according to claim 3, characterized in that: The method of identifying a dialogue application scenario based on the spoken dialogue speech and the dialogue knowledge database comprises the following steps: Perform semantic recognition on the actual dialog box of the spoken conversation speech to obtain the semantics of the actual dialog box; When the actual dialog box semantics of two actual dialog boxes are related, a corresponding relationship is established between the two actual dialog boxes; At least one actual dialogue set is formed, the actual dialogue set is composed of actual dialog boxes, and any two actual dialog boxes in the actual dialogue set are in a corresponding relationship; Obtaining the associated conversation set with the largest overlap with the actual conversation set, matching it to the actual conversation set, and matching the occurrence context of the associated conversation set to the corresponding actual conversation set; The occurrence situations corresponding to the actual dialogue set of spoken dialogue voice are deduplicated and summarized to obtain the dialogue application scenario.

5. The method for evaluating the quality of oral dialogue for college English learning according to claim 4, characterized in that: The identification of unreasonable content and obtaining the semantic misunderstanding part includes the following steps: The associated dialogue set corresponding to the occurrence scenario with the greatest overlap with the dialogue application scenario is used as the preliminary associated dialogue set; Taking one of the at least one actual dialog box as the target actual dialog box; The actual dialog set containing the target actual dialog box is used as the target actual dialog set; The preliminary associated dialogue set, whose number of elements is consistent with the target actual dialogue set, is used as the secondary associated dialogue set; The secondary associated conversation set with the largest overlap with the target actual conversation set is used as the target associated conversation set; Arrange the sample dialog boxes in the target associated dialog box set in order to obtain a first sequence, and arrange the actual dialog boxes in the target actual dialog box set in order to obtain a second sequence; Establish a correspondence between the sample dialog boxes and the actual dialog boxes at the same position in the first sequence and the second sequence, and use the sample dialog box corresponding to the target actual dialog box as the target sample dialog box; Obtain the part of the target actual dialog box that is different from the target sample dialog box as the suspicious part; Based on big data, we can obtain commonly used keywords in spoken language and classify them. Keywords of the same type have the same semantics. Use the same type of keywords to replace the suspicious part, and obtain all replacement situations after the suspicious part is replaced; In each replacement case, the part of the target actual dialog box that is different from the target sample dialog box is retrieved as the unreasonable part, and the part of the target actual dialog box that is the same as the target sample dialog box is retrieved as the reasonable part; The unreasonable part with the smallest length is regarded as the semantic misunderstanding part; When the target actual dialog box traverses at least one actual dialog box, at least one semantically misunderstood part is obtained.

6. A method for evaluating the quality of oral dialogue for college English learning according to claim 5, characterized in that: The step of analyzing the unreasonable degree of the semantic misunderstanding part to obtain the misunderstanding degree coefficient comprises the following steps: In the target sample dialog box, the reasonable part corresponding to the semantically misunderstood part is removed to obtain the reference part; Based on big data, the probability of two keywords appearing simultaneously is counted as the correlation between the two keywords; Acquire all possible combinations of keywords in the semantic misunderstanding part and keywords in the reference part to obtain at least one keyword group; The reciprocal of the relevance of the keywords in at least one keyword group is taken as the average to obtain the misunderstanding degree coefficient.

7. A method for evaluating the quality of oral dialogue for college English learning according to claim 6, characterized in that: The analysis to obtain the pronunciation problem degree and initiation response speed of the actual dialog box comprises the following steps: Compare the standard pronunciation of English words with the sample conversation content to obtain at least one sample pronunciation; Count the minimum time interval between adjacent sample pronunciations in chronological order as the feature time; Recognize the pronunciation interval of the content in the actual dialog box, and use the point where the pronunciation interval time is greater than the characteristic time as the segmentation point; The content in the actual dialog box between adjacent segmentation points is used as the actual word pronunciation; Obtain words whose pronunciation differs from the actual word pronunciation by less than a preset value as suspected words, summarize the suspected word set, and use the suspected word with the smallest pronunciation difference from the actual word in the suspected word set as the target word, and use the target word to replace the position of the actual word pronunciation in the actual dialog box to obtain the actual change dialog box, and the preset value is a parameter set based on the experience value; When the semantics of the actual change dialog box is consistent with that of the actual dialog box, the difference between the standard pronunciation of the target word and the actual word pronunciation is counted as the degree of the local problem. Otherwise, the target word is deleted from the suspected word set and the previous step is repeated until the degree of the local problem is counted. The local problem degrees are accumulated to obtain the pronunciation problem degree; The time intervals between adjacent actual dialog boxes are averaged to obtain the initiation reaction speed.

8. The method for evaluating the quality of oral dialogue for college English learning according to claim 7, characterized in that: The analysis based on the proportion of semantic misunderstanding to comprehensively obtain the dialogue coherence index includes the following steps: The proportion of the semantic misunderstanding part in the spoken conversation speech is multiplied and added with the misunderstanding degree coefficient of the semantic misunderstanding part to obtain the conversation coherence index.

9. A method for evaluating the quality of oral dialogue for college English learning according to claim 8, characterized in that: The weighting of the pronunciation problem degree, the initiation response speed and the conversation coherence index comprises the following steps: Based on the analytic hierarchy process, weights for the degree of pronunciation problems, speed of initiation response, and conversation coherence index were formed.

10. The method for evaluating the quality of oral dialogue for college English learning according to claim 9, characterized in that: The method of obtaining a quality assessment value range using the sample conversation content comprises the following steps: Obtaining a quality evaluation value of the sample conversation content in the same manner as the quality evaluation value of the spoken conversation speech; The maximum and minimum values ​​of the quality evaluation values ​​of the sample conversation contents are used as endpoints to form a quality evaluation value range.

Images