TWI575483B - A system, a method and a computer programming product for learning? foreign language speaking - Google Patents

Description

Foreign language spoken learning system, method and computer program product

The present invention relates to a language learning system, method and computer program product, and more particularly to a system, method and computer program product for foreign language speaking and learning.

Foreign language mouth training has a lot of related computer-aided learning applications, but each has its own shortcomings. Take the English speaking learning system as an example. The existing learning methods can usually be divided into text reading and conversation response. In the text-reading learning mode, the learning system (such as MyET, English fluency, SpeakingPal, etc.) displays the English sentence in words, allowing the user to read the sentence, and then the system evaluates the correct reading. Sex, which sometimes provides ratings for certain pronunciation indicators. But this type of learning, from the four aspects of listening, speaking, reading, and writing in language learning, is actually practicing "reading" that belongs to the input of ideas, rather than practicing "speaking" that belongs to the output of ideas. It is conceivable that even if users use such systems to practice foreign language reading and pronunciation very standard, but because they have not practiced to use English to express their ideas, when they encounter foreign language, they are still unable to do anything.

In the conversation-reactive learning mode, the learning system (such as ESL Robot Pro, Pandorabots CallMom, etc.) allows users and the system to use text or voice freedom. chat. However, this type of learning system only responds to the sentences input by the user, rather than actively guiding the teaching, and also lacks the supporting course materials for users to study and practice. The result is that users often don't know what to say to the blank dialog box, and they don't know how to start learning.

In view of this, the present invention provides an improved foreign language speaking learning system, thereby solving the problems that the conventional teaching system cannot solve.

The invention integrates the speech recognition (Chatbot) technology in the field of Speech Recognition and Artificial Intelligence, and constructs a foreign language speaking teaching system with a virtual dialogue exercise environment. The system consists of the following main components: a user module, which is responsible for the operation of the host user, and also presents the message output by a system to the user in a voice and text manner, and is responsible for receiving the voice response sent by the user and transferring Text input; a system master module, responsible for the master server operation to serve at least one user, and to lead and control interactive dialogue learning with the user, that is, output a message to the user according to the course content, And comparing whether the response input of the user meets a predetermined response; wherein the system master module can realize a dialogue interaction between the system and the user by using a dialog robot module, or is controlled by the system The module writes the code to achieve the related functions; and a set of interactive oral learning materials, which are designed for the interaction between the system and the user, and supports the system to provide at least one interactive learning mode.

Through the integration operation between the above components, the present invention allows the user to conduct a human-machine dialogue exercise with the virtual teacher and the dialogue object played by the system in a spoken manner, simulating the interaction between the actual human and the human. its In the present invention, the functions of the teacher's guidance, questions, answers, prompts, and interaction between the students in the physical classroom teaching are handed over to the system with intelligent interactive characteristics, thereby developing an innovative interactive learning mode. .

One of the features of the present invention, the interactive learning mode includes a role dialogue mode, in which the system and the user respectively play two dialogue roles of a thematic dialogue script in the interactive oral learning textbook. In order to conduct a speaking dialogue exercise; wherein, the system will speak according to a statement of the dialog script, and compare whether the user's response to the statement meets the predetermined response statement, and if not, the system will request to use The person responds again. If it is met, the system will make a statement of the next statement of the dialogue script, and this will continue until all the statements in the dialog script have been completed. Here, combined with the interactive scripting material research and oral dialogue interaction, the user will study and familiarize with the vocabulary and sentence meaning in the dialogue script in a study mode, and then use the role dialogue mode and the system to talk about the script content. In the dialogue drill that talks about conversation, it is necessary to tell the correct role statement through speech recognition and let the system understand and interact. In addition, in order to prevent the user from interrupting the conversation during the conversation without knowing how to respond to the next sentence, the present invention will provide a mechanism for prompting, as will be explained later.

According to another feature of the present invention, the interactive learning mode also includes a request response mode. In this mode, the system will present a request, a question or a setting status for the exercise content, and the user will respond with a voice. For example, in a sentence-type sentence-making exercise, part of the sentence words can be proposed by the system, and the user can respond to the complete statement. For example, various idioms that express gratitude or apology can be made by the system to make a situation, and the user exercises the various idioms provided by the course to form a response, such as the interactive effect of group participants in the physical classroom. . Here, the request response mode is not limited to one question and one answer, but the process can have multiple dialogue interactions. In addition, the request response mode will make good use of the characteristics of the computer system that is easy to repeat actions, and provide a practice example in which a plurality of contents are randomly changed. The problem is to allow the user to perform a large number of repetitive exercises to achieve proficiency in other statements that are difficult to achieve in other teaching systems. Similarly, in order to prevent the user from continuing to respond during the conversation without knowing how to respond, the present invention will provide a mechanism for prompting, as explained below.

The second feature of the present invention is that in order to prevent the user from reading the missing words in a word-by-word manner, but not detracting from the guidance required by the user in the dialogue interaction, the system does not display the statement to be responded in plain text in the dialogue interaction. The user reads, but provides a prompt instruction. When the system receives the prompt instruction, a prompt will be presented in the manner of voice plus short text flashing assistance. The prompts may be plural, and the user may use the prompt instructions one by one. In this way, when the user does not know how to respond in the conversation interaction, the prompt instruction can be used to prompt, and the conversation interaction can be continued without interruption. At the same time, because the prompt text only flashes briefly, the user does not have the opportunity to read the text, but the user must repeat it after listening to the prompt, which is a true oral expression, so the user will be immersed in the real In the atmosphere of listening and speaking, you can effectively exercise and improve your speaking and speaking skills.

Existing systems usually only provide a single fixed dialogue script, but the actual daily conversation is very flexible. The same sentence can be used in a variety of ways. Therefore, it provides a variety of ways to simulate the daily conversation situation and avoid the boring user experience. The content of the dialog script for a single fixed statement is necessary. Herein, another feature of the present invention is that any statement in a dialog script in an interactive oral learning textbook is composed of at least one synonym sentence, each synonym sentence has the same semantic but is used in word selection and sentence pattern. Different statements that change in tense, active, passive, and other language elements. In this way, the various dialog scripts provided by the system can not only meet the above-mentioned diverse statement requirements, but also allow the user to contact a large number of instances of the multi-language elements covered by the synonymous sentences, thereby obtaining a reference learning opportunity that is multiplied by a single fixed script of the sentence sentence, thereby accelerating the enhancement. His mouth speaks expressive ability.

Accordingly, the present invention has another feature. When the system presents a dialog script for the user to study, a synonym sentence transformation instruction is provided. Upon receiving the instruction, the system will perform synonym replacement on the dialog script, that is, random. Change at least one statement in the dialog script to a synonym of the statement. At this time, the dialog script will present another look of semantics and context, so that the user can easily learn diverse and changeable. Conversational content combination.

Another feature related to the present invention is that in the role dialogue mode, the system's speech is selected from the synonymous sentence of the dialog script, and the user's response can be any synonymous sentence of the response statement. At this time, at least one of the synonymous sentences will also become at least one prompt for the user to access. In this way, the user will be able to practice the interactive use of the synonymous sentences in the dialogue interaction, simulate the diversification of the oral expression in the daily conversation, and expand the practice of using more everyday idioms.

A second feature of the present invention is to allow the user's response to be flexible in the sense of semantics without having to be identical to the predetermined response statement in order to conform to the daily conversation reality and to create a flexible and interesting user experience. . Such statement expression flexibility can be solved by the code inside the system main control module, and can also be easily and conveniently achieved by using off-the-shelf software components such as a dialogue robot module. If the dialog robot module is used, the elasticity is realized by using a scripting language of the dialog robot to write a statement pattern with sentence expression flexibility, and the expression expressiveness includes synonymous words. Replace the substitution, the existence of non-keywords (such as adverbs, adjectives, and pronouns), and the choice of the position of the adverbs (see the examples in the implementation section for details) to form the required statement flexibility. Accordingly, the present invention provides a user's response to the expressiveness that is not available in the prior art to create a smart and interesting experience that approximates everyday conversations.

According to still another feature of the present invention, the interactive oral speaking learning material of the present invention includes a test, and at the same time, in order to enhance the effectiveness of the oral speaking exercise and improve the interactive interest of the test, the test system is conducted in a spoken manner in the present invention. . Under the spoken test, the system will present the question in both voice and text, and the user will respond in a verbal manner to answer, and the system will determine whether the response meets the predetermined response. At this time, whether it is an open question-and-answer or a multiple-choice test, the system can provide a question prompt by prompting the command; if it is a multiple-choice test, the system will include the answer option, or the answer option will prompt the command. Way to present. In addition, the user's answer can also have the statement expression flexibility without having to change the statement with the predetermined answer.

Another object of the present invention is to provide a foreign language spoken language learning method, which is implemented by the aforementioned foreign language spoken learning system, wherein the method comprises the following steps: the system master module controls and controls the system and the user Inter-learning interactive learning, and outputting a message to a user module according to the course content; the user module outputs the message to the user in at least one of voice and text, and receives the user from the use The voice response of the person and the voice response are converted into a text input response; the text input response is received by the system master module, and the comparison is judged whether it meets a predetermined response. It can be seen that the method allows the user's voice response to judge the correctness through the system, and accordingly, the dialogue response is performed, just as the user talks with the virtual dialogue role played by the system. Thereby, the user can perform a dialogue exercise simulating an actual person-to-person dialogue mode.

1‧‧‧ Foreign language speaking learning system

2‧‧‧User Module

3‧‧‧System Master Module

4‧‧‧Dialog Robot Module

5‧‧‧Interactive oral learning materials

20‧‧‧User interface

21‧‧‧Voice recognition unit

22‧‧‧Text-to-speech unit

30‧‧‧Main system functional unit

31‧‧‧Course Service Unit

33‧‧‧Dialog robot interface unit

40‧‧‧Dialog robot control script

230‧‧‧Dialog script

230’‧‧‧Dialog script

300‧‧ ‧ courses

310‧‧‧Dialog script

311A, 321A‧‧‧ Synonyms

312A, 322A‧‧‧ Synonyms

311B‧‧‧Dialog statement

301‧‧‧Play button

302‧‧‧ statement conversion button

401‧‧‧Robot speaking display area

402‧‧‧Microphone start command

403‧‧‧User voice display area

404‧‧‧ prompt instructions

405‧‧‧Users speak input instructions

406‧‧‧ prompt display area

433‧‧‧ voice messages

434‧‧‧ text input

51‧‧‧Scheduled response statement

52‧‧‧Flexible statement

61‧‧‧ Practice start message

62‧‧‧Sentence exercises

63‧‧‧Correctly answered keyword words

64‧‧‧Respond correctly

65‧‧‧ Practice start message

66‧‧‧ Greetings

67‧‧‧ idioms

68‧‧‧ Keywords

69‧‧‧Complete response statement

S71~S79‧‧‧Steps

1 is a schematic diagram of a system architecture of an embodiment of a foreign language spoken learning system 1 of the present invention.

Figure 2 (A) is a schematic diagram of a dialogue script in one of the courses.

Figure 2 (B) is a schematic diagram of a synonymous version of the dialog script.

Fig. 3(A) is a schematic diagram showing a dialog script displayed in the study mode.

Fig. 3(B) is a schematic diagram of a synonym sentence transformation instruction executed in the study mode.

Figure 4 (A) is a first state diagram of one of the character dialogue modes.

Figure 4 (B) is a second state diagram of one of the character dialogue modes.

Figure 4 (C) is a third state diagram of one of the character dialogue modes.

Figure 5 is a schematic diagram of a predetermined response statement and its statement expression elasticity.

Figure 6 (A) is a schematic diagram of the operation of one of the sentence practice modes.

Figure 6 (B) is a schematic diagram of one of the commonly used expression syntax practice modes.

FIG. 7 is a schematic flow chart showing one main steps of a foreign language spoken language learning method of the present invention.

1 is a schematic diagram of a system architecture of an embodiment of a foreign language spoken learning system 1 of the present invention. In this embodiment, the foreign language speaking learning system 1 utilizes a dialog robot module 4 to perform a function of comparing and determining a user response sentence, which may include a user module 2 and a system main control module 3. The dialogue robot module 4 and an interactive spoken language learning material 5. The user module 2 is disposed on an electronic device and can transmit data or instructions to the system main control module 3. The system main control module 3 is disposed on a first network server for receiving and processing data or instructions from the user module 2, and the system main control module 3 can also transmit data or instructions. To the dialogue robot module 4 or the user module 2. The dialog robot module 4 is disposed on a second network server for receiving and processing data or commands from the system master module 3, and can transmit data or commands to the system master module 3 . Wherein, the first network server and the second network server can be two different network servers, but To be the same network server. In addition, the electronic device, the first network server and the second network server are all connected to the Internet device, so that the foreign language port said that the modules in the learning system 1 can communicate with each other. It should be noted that in other embodiments, the function of the dialog robot module 4 can also be achieved by writing a large amount of code in the system main control module 3.

Preferably, the electronic device is a mobile electronic device, such as a smart phone, a tablet computer or a portable device capable of executing an application, but it can also be a non-mobile electronic device having an executable application such as a desktop computer. Device. The user module 2 can be a computer program product loaded on the electronic device, and a user interface 20 is provided on a display screen of the electronic device to interact with the user according to the user input. Instructions to operate. Preferably, the user module 2 is an application software installed on a mobile device, such as an application of a currently popular smart phone or tablet.

The user module 2 can receive a voice message spoken by a user interface through a microphone device. The microphone device can be an element on the electronic device or an external device connected to the electronic device. After receiving the voice message, the user module 2 can convert the voice message into a text input by using a voice recognition unit 21, and the voice recognition unit 21 can be a voice recognition function built in the electronic device. It can be a component of the user module 2 itself. Further, the user module 2 transmits the text input to the system main control module 3 for subsequent processing. In addition, the user module 2 can also convert a text message from the system main control module 3 into a voice message by using a text-to-speech unit 22, and by using a speaker device or the electronic device on the electronic device A connected speaker device is output. Similarly, the text-to-speech unit 22 can be a text-to-speech function built into the electronic device, or can be a component of the user module 2 itself.

The system main control module 3 can be a processing system mounted on the first network server, and includes a main system function unit 30 for controlling the operation of the system main control module 3. Main system The functional unit 30 is linked to a course service unit 31, or the main system function unit 30 includes the course service unit 31. The course service unit 31 can obtain the interactive speaking learning material 5 from a database on the first network server. In addition, the database can also exist on another database server.

The main system function unit 30 is also coupled to a dialog robot interface unit 33 or may itself comprise the dialog robot interface unit 33. The dialog robot interface unit can transmit instructions or data to the dialog robot module 4, and can also receive data or instructions from the dialog robot module 4.

Preferably, the above unit is a plurality of programs on the first network server for executing, so that the main system function unit 30 can perform the foregoing functions, but is not limited thereto.

The dialog robot module 4 can be any existing dialog robot system, such as Alicebot, Pandorabots, CleverScript, etc., and operates according to a dialog robot script (40) written to analyze the text from the user. Whether the input response conforms to a statement style. If not, the dialog robot module 4 will notify the user to retry through the system master module 3; if it is met, the dialog robot control script 40 will follow the dialog. dialogue. Further, the dialogue robot control script 40 corresponds to the interactive speaking learning teaching material 5, whereby the foreign language speaking learning system 1 can perform the behavior of the virtual teacher as described above by the dialogue robot module 4.

The interactive oral learning teaching material 5 is an example of providing English learning, but in fact, the present invention can also be applied to other languages, as long as the interactive oral speaking learning material 5 of the language is replaced in the implementation, and the language is applied. The robot module and related control scripts are all available. In order to make the description more detailed, the following is an example of English.

The interactive speaking learning teaching material 5 is stored in the database on the first web server in the form of text data, and has the interactive interaction characteristic between the learning system 1 and the user for the foreign language. Design a plurality of courses and support the foreign language speaking learning system 1 to provide at least one interactive learning mode. Each course can include dialogue scripts, relevant textual textbooks that can be read by users, or other forms of textbooks such as sentence-making exercises. When the interactive oral learning learning material 5 course is transmitted to the user module 2, the nature of the content may be directly displayed on the user interface 20 in text form, or via the text-to-speech unit 22 Converted to a voice data, and then output by the speaker device.

In an embodiment, after the user selects a course, the user module 2 sends a course request message to the system main control module 3, so that the main system function unit 30 learns from the interactive port. The content of the course is obtained in the teaching material 5 and transmitted to the user module 2 for display on the user interface 20.

Figure 2 (A) is a schematic diagram of one of the dialog scripts 230 of one of the courses. In this example, the dialog script 230 is composed of a plurality of dialog statements, and includes at least two dialog characters A and B, each of which has a plurality of sentences and is continuous and mutually In response to the conversation. The content of the dialog script 230 is merely an example and is not a limitation.

Preferably, each dialog statement consists of at least one synonym. As shown in FIG. 2(B), if a partial dialog sentence in the dialog script 230 contains two synonymous sentences, the dialog script 230 conceptually becomes another dialog script 230', in which two synonymous sentences separated by a slash are separated. One can be taken at will without changing the semantics and context of the entire conversation.

According to the foregoing, the design of the interactive oral learning teaching material 5 supports the foreign language speaking learning system 1 to provide a plurality of interactive learning modes, and the following will be described for the plural interactive learning modes.

First, a study mode will be described, and please refer to FIG. 1 and FIG. 3(A) to (B) at the same time. FIG. 3(A) is a schematic diagram showing a dialog script 310 of a course 300 displayed on the user interface 20 in the study mode. In the study mode, the user can use a play button 301 to listen to a plurality of dialog statements that learn the dialog script 310 to familiarize themselves with the content of the dialog. At this time, the foreign language speaking learning system 1 can also provide functions such as a green word query and a grammar commentary to assist the user in learning. The content of the dialog script 310 is merely an example and is not limited.

Moreover, in the study mode, the dialog script 310 can perform synonymous sentence transformation by a synonym sentence transformation instruction 302. As shown in FIG. 3(B), when the synonym sentence conversion instruction 302 is executed by the user (that is, when the user module 2 receives the synonym sentence transformation instruction 302), at least one sentence 311A is replaced with its synonym. The sentence 321A causes the dialog script 310 to be transformed into a different version 320 of semantics and context unchanged.

After the user is fully familiar with the content of the course script 310 in the study mode, the user can enter the role dialogue mode to perform a dialogue interaction exercise with the dialogue robot module 4. The character dialogue mode will be described below, and please refer to FIG. 1 , FIG. 3 (A) to (B), and FIGS. 4 (A) to 4 (C).

In the character dialogue mode, the user acts as one of the dialog characters in the dialog script 310, and the other dialog character is played by the dialog robot module 4 and performs a dialogue. For example, if the dialog robot module 4 plays the dialog character A in the dialog script 310 and speaks a dialog statement (for example, 311A), the user plays the role B, and must say the phase. Corresponding to a predetermined response (eg 312A). The dialogue characters played by the user and the dialogue robot 4 can also be interchanged.

In addition, the dialog robot module 4 selects a sentence from at least one synonym of a dialog statement of the dialog script 310 according to the dialog robot control script 40 (eg, selects one of the synonym sentences 311A and 321A). ). Thereafter, the user must speak a response by the microphone device, and when the dialog robot module 4 compares and determines that the user's response conforms to a statement style of a predetermined response (eg, 312A or 322A), The dialog robot module 4 will then speak the next statement (e.g., 311B) until the dialog script 310 ends.

FIG. 4(A) is a schematic diagram showing a first state of the user interface 20 in the character dialogue mode, the first state being a state when the dialog robot module 4 speaks. In this case, the user interface 20 preferably includes a robot speech display area 401, a microphone activation command 402, a user speech display area 403, a prompt command 404, a user speech input command 405, and a prompt display. Zone 406, but not limited.

The robot speech display area 401 is configured to display the robot speech statement (eg, one of the equivalent sentences 311A, 321A), and the microphone activation command 402 is used to activate the microphone device to enable the user to start voice input. The user module 2 can display the robot speech statement in the form of text in the robot speech display area 401, or can be converted into a voice form by the text-to-speech unit 22, and then played by the speaker device, and Or you can present it both in text and in speech. When it is the turn of the user to speak, the user must press the microphone activation button 402 to give a correct response.

FIG. 4(B) is a second state diagram of the user interface 20 in the role dialogue mode, and the second state refers to a state when the user responds. When the user utters a response (ie, generates a voice message 433), the voice message 433 is converted to a text input 434 via the voice recognition unit 21 and displayed in the user speech display area 403. At this point the user can know their own Whether the mouth is correctly identified. If there is a gap in the identification result, the user can press the microphone activation command 402 to perform another mouth-speaking, or change the keyboard with the keyboard connected to the electronic device for input correction. After the user is satisfied with the content of the user speaking on the display area 403, the user can press the user input command 405 to cause the text input 434 to be transmitted to the system main control module 3, and then transmitted to the conversation. The robot module 4, based on the dialog robot control script 40, determines whether the text input 434 conforms to the statement pattern of its predetermined response statement (e.g., dialog statement 312A). If yes, the dialog robot module 4 sends a next sentence of the robot dialog statement (for example, the dialog statement 311B) according to the setting of the dialog robot control script 40 to continue the dialogue; if not, the dialog robot module 4 will send a speech. Correct and ask the user to retry the message. This is repeated until the interactive dialogue exercise of the entire dialogue script 310 is completed.

FIG. 4(C) is a third state diagram of the user interface in the role dialogue mode, and the third state refers to a state when the user executes the prompting instruction 404. When the user executes the prompting instruction 404 (that is, when the user module 21 receives the prompting instruction 404), the user module 2 will briefly flash a prompt in the prompt display area 406. Or play the prompt through the speaker device, or flash and play briefly at the same time. In other words, when the user does not know how to respond, just press the prompt to press the command 404, and then the prompt is obtained. Wherein, the prompt may be a complete sentence, a part of a word or a similar sentence of a predetermined response sentence (for example, 312A or 322A).

In an embodiment, if the predetermined response has a plurality of prompts, when the user executes the prompting instruction 404 again, the user module 2 provides another prompt to the user. For example, when the user presses the prompt command 404, the user module 2 provides a synonym 312A of the predetermined response to the user. When the user presses the prompt command 404 again, the user The user module 2 will provide another synonym 322A to the user.

In addition, the present invention does not limit the response of the user to only the response statement 312A. By means of the settings in the dialog robot control script 40, the equivalent sentences 312A and 322A can be accepted by the dialog robot module 4. In other words, as long as the user speaks one of the equivalent sentences 312A and 322A, the dialog robot module 4 determines that the user's response is correct.

In another preferred embodiment, the dialog robot control script 40 supports the setting of statement expression flexibility, and the response statement of the user may have elasticity without changing the semantics without the predetermined response statement (for example, the One of the equivalent sentences 312A and 322A) is identical. The following describes how the dialog robot module 4 achieves the features of the preferred embodiment, and please refer to FIG. 1 and FIG. 5 at the same time.

Figure 5 is a schematic diagram of a predetermined response statement and its flexible statement. When the content of a robot statement 50 is "A saving account will be good for you.", the content of a predetermined response statement 51 is "Well, what is the interest rate for this type of account?", if allowed at this time The various expressions of elasticity described by the flexible statement 52, the user only needs to say any one of the sentences, the dialogue robot module 4 will judge the correct response. Moreover, the above statements are merely examples and are not limiting.

As mentioned above, such statement expression flexibility is achieved by using a dialog robot scripting language to write statement styles that are more flexible.

The following describes the embodiment of the statement style. Since the scripting language and its writing and writing are extremely complicated and complicated, it cannot be elaborated here, and only a high-level introduction can be made. For the flexible statement 52 in FIG. 5, taking the scripting language of the dialog robot ChatScript as an example, the expression pattern expressing the elasticity will be "({well}{,}[(what{will}be)({~can_you}tell me)) (how about)]*1 interest rate for[such(this type of)]account)". Among them, the innermost brackets are the style content, the braces {...} represent the optional elements, such as well is optional; the brackets [...] represent the elements (including the parentheses) Included), such as [such (this type of)] means such or this type of The sign at the beginning of the symbol ~ represents any element of a set. For example, the content of the ~can_you collection is set to can you and could you please; the symbol *1 indicates any word, for example, the or any. This set of statement styles will enable the dialog robot module to match any of the statements in the flexible statement 52, resulting in the required statement flexibility.

It is worth noting that the operation and function of the dialogue robot module 4 in the role dialogue mode and the statement expression flexibility can also be achieved by writing a large amount of code in the system main control module 3.

Next, the present invention will be described with respect to the request response mode, and please refer to FIG. 1 and FIG. 6(A) to (B) at the same time.

FIG. 6(A) shows the operation of a sentence sentence practice example. The dialog robot module 4 can first provide a practice start message 61 to the user through the user interface 20, prompting the upcoming sentence sentence composition practice content. . Thereafter, the dialog robot module 4 provides a sentence practice title 62 containing a question and a correctly answered keyword word 63, and the user must utter the answer by the keyword word 63. A complete sentence sentence is composed, and the dialog robot module 4 further determines whether the user's answer is correct by the dialog robot control script 40. When the user's answer is incorrect, the dialog robot module 40 will ask the user to re-answer; and when the user speaks the correct answer 64, the dialog robot module 4 will provide the next sentence-making exercise topic again. Give the user a repeat of the practice. If the user does not know how to answer correctly at any time, use the prompt command to get a prompt. In addition, the foreign language speaking learning system 1 can provide a sufficient number of problem sets, and each time selects the required number to perform the exercises in a random and indefinite order to create a smart and interesting learning experience.

Figure 6(B) illustrates the operation of a idiom practice seeking help. The dialog robot module 4 provides a practice start message 65 to the user, prompting the upcoming idiom expression practice content. Thereafter, the dialog robot module 4 initiates an interactive process that is repeated several times, beginning with a dozen greetings 66, the user having to seek help using one of the idioms 67 taught in a course, the dialog robot module 4 then Responding to and prompting a keyword 68 for seeking help content, the user must now use the keyword 68 to correctly formulate a complete response statement 69 for the desired help content. In the process, the dialogue robot module 4 still determines whether the user's response is correct by the dialogue robot control script 40. When the user's answer is incorrect, the dialog robot module 40 will ask the user to re-answer, and if the answer is correct, the dialog robot module 4 can re-enter the next question for practice. Similarly, if the user does not know how to answer correctly at any time during the conversation, the prompt instruction can be used to obtain the prompt; in addition, the foreign language speaking learning system 1 can prepare a sufficient number of keyword sets, which are randomly provided each time to make the animation interesting. Learning experience.

It is worth noting that the operation and function of the dialog robot module 4 in the response mode can also be achieved by writing a large amount of code in the system master module 3.

Next, the present invention will be described with respect to the one-stop test mode. The oral test mode can be divided into at least a multiple choice question test and an open question and answer test.

In an embodiment of the multiple-choice test, the dialog robot module 4 provides a test question and a plurality of answer options through the user interface 20 and the speaker device, with voice and text display, including For example, using the prompting instruction as described above to present the plurality of answer options, when the user selects one of the answer options, the answer option is verbally answered through the microphone device, and the dialog robot module 4 can The dialogue robot control script 40 determines whether the user's answer is correct. Thereby, the user's answer can have the expression flexibility as described above.

The open question and answer test is similar to the multiple choice test, but there is no answer option. Under the open question and answer test, the dialogue robot 4 transmits through the user interface 20 and the speaker device. The voice can be accompanied by a text display to ask a question, and the user can receive an answer through the microphone device, and can provide a prompt by the prompt instruction of the user interface, after which the dialog robot module 4 can The dialog robot controls the script 40 to determine if the user's answer is correct. Thereby, the user's answer can have the expression flexibility as described above.

Therefore, the dialogue robot module 4 uses the voice method to present the test question, and the user must also answer the question in a spoken manner, so the user can not only learn his own learning under the test. Results, but also extra training to the ability to speak and speaking.

It is worth noting that the operation and function of the dialog robot module 4 in the talk mode can also be achieved by writing a large amount of code in the system master module 3.

FIG. 7 is a flow chart showing the main steps of an embodiment of a foreign language spoken learning method according to the present invention, which is executed by the foreign language speaking learning system 1 described above. The method includes the following main steps: first, step S71 is performed, through the system main control module 3, the dialogue robot module 4 serves as a virtual teacher and a dialogue object role, to learn the teaching material 5 according to the interactive mouth. a course content outputting a message; then proceeding to step S72, the system master module 3 processes the message sent by the dialog robot module 4, and transmits the message to the user module 2; Step S73: displaying the message to a user through the user interface 20, or converting the message to a voice through the text-to-speech unit 22, playing the voice through the speaker device, or simultaneously displaying the message and playing The voice is sent to the user module 2 through the microphone device to receive a response voice of the user; and then the step S75 is performed by the user module 2 through the voice recognition unit 21 The response voice is converted into a text input response; then step S76 is performed, the system main control module 3 receives the text input response, and the text input response The dialog module 4 to the robot; After step S77, the robot by the session module 4 determines the character of the conversation in accordance with the robot control script 40 Entering whether the response conforms to a statement style of a predetermined response; if yes, proceeding to step S78, by which the dialog robot module 4 provides the next message according to the course content, and restarts execution from step S72 until the end of the course; If yes, step S79 is performed, and the dialog robot module 4 transmits a retry message to the user module 2 through the system main control module 3 to request the user to respond again, and then re-steps from step S73. Begin execution. It should be noted that the steps of the dialog robot module 4 in the above main steps can also be achieved by writing a large amount of code in the system master module 3.

Preferably, the preset style has the flexibility of comparison as described above, so that the foreign language speaking learning system can make the response of the user flexible after executing the method.

Based on the above main steps, and further incorporating more detailed steps, the foreign language speaking learning system will be able to provide such learning modes as described above.

In an embodiment, the foreign language speaking learning method further includes the steps of: by the dialogue robot module 4, playing the two dialogue roles of the dialogue script in the interactive speaking teaching material 5 with the user respectively, and performing a dialog drill to provide the role dialogue mode; wherein the play is performed by the dialog robot module 4 according to a statement of the dialog script, and the comparison determines whether the response of the user conforms to the statement The statement style of a predetermined response statement, if not met, the user is required to respond again, and if it is met, the speech of the dialog script is executed. By further performing this step, the foreign language speaking learning system 1 can provide the role dialogue mode to the user, and can also generate a synonymous sentence as the speech is randomly selected by the dialogue robot 4, and the dialogue robot 4 can The judgment of the user's response can be characterized by flexibility. Since the relevant content of the character dialogue mode has been described in detail in the previous paragraph, it will not be described in detail here.

In another embodiment, the foreign language speaking learning method further comprises the steps of: speaking, by the dialogue robot module 4, to present a request message to the user, and by using the dialogue robot module 4 Receiving a response of the user in a spoken manner, and comparing a statement style of determining whether the response meets a predetermined response of the request message; wherein, if not, the user is speaking by the dialog robot module 4 Re-respond to provide the response model for this request. By performing this step, the foreign language speaking learning system 1 will be able to provide the aforementioned response response pattern to the user and produce the effects as described above. Since the relevant content of the response mode has been described in detail in the previous paragraph, it will not be described in detail here.

In another embodiment, the foreign language speaking learning method further includes: when the user module 2 receives a prompting instruction through the user interface 20, providing at least one prompt; wherein the user module is 2 Convert the prompt to voice to play, or the prompt may be briefly flashed on the user interface 20, or may be played in both voice and text. By performing these steps, the foreign language speaking learning system 1 will be able to provide the prompting function as described above to the user. Since the relevant content of the prompt function has been described in detail in the previous paragraph, it will not be described in detail here.

In an embodiment, the foreign language speaking learning method further includes: when the user module 2 receives a synonym sentence conversion instruction, the user module 2 displays the user interface 20 The dialog script performs synonymous sentence substitution to randomly change at least one statement in the dialog script to a synonymous sentence of the statement. By performing these steps, the foreign language speaking learning system 1 can provide various synonymous sentences of the dialog script and the dialog sentence to the user for learning. Since the relevant content of the synonymous sentence replacement has been described in detail in the previous paragraph, it will not be described in detail here.

In another embodiment, the foreign language speaking learning method further includes: presenting a test question by the dialogue robot module 4; and displaying the test question by voice and supplemented by the user module 2, and Receiving a response from the user in a spoken manner; and determining whether the response is correct by the dialog robot module 4 in the preset style to provide the spoken test mode. Thereby, the The foreign language speaking learning system 1 can provide the selective speaking test mode such as the aforementioned multiple-choice test or the open question and answer test to the user. Since the relevant content of the spoken test mode has been described in detail in the previous paragraph, it will not be described in detail herein.

It should be noted that the steps of the dialog robot module 4 in the steps of achieving various interactive modes can also be achieved by writing a large amount of code in the system master module 3.

By means of the foreign language spoken learning system and the foreign language spoken learning method of the present invention, when the user learns a dialog script, the user can learn a plurality of synonymous sentences of the dialog statement, which can be learned more than the traditional sentence fixed dialogue script. Rich fun and learning efficiency. In addition, when the user encounters a situation in which the user does not know how to answer during the dialogue with the dialogue robot, the present invention also adopts a voice and text flashing prompting manner to induce the user to express a response sentence to enhance the effectiveness of the oral practice. . At the same time, the user does not need to use the same statement as the dialog statement in the dialog script to respond to the dialogue. Instead, the syntactic sentence can be selected with any syntactic sentence and the statement configuration to form a response statement, which makes the practice of the mouth more close. Dialogue in real life, and not boring. Furthermore, the execution of the internal test of the present invention is also carried out by means of oral dialogue, which further increases the chance of the user exercising the speaking ability. In summary, the present invention will be able to break through the limitations of prior art and existing products, and provide a foreign language spoken learning environment with high interactivity, high efficiency, and even high interest.

The above-mentioned embodiments are merely examples for convenience of description, and the scope of the claims is intended to be limited to the above embodiments.

1‧‧‧ Foreign language speaking learning system

2‧‧‧User Module

20‧‧‧User interface

21‧‧‧Voice recognition unit

22‧‧‧Text-to-speech unit

3‧‧‧System Master Module

30‧‧‧Main system functional unit

31‧‧‧Course Service Unit

33‧‧‧Dialog robot interface unit

4‧‧‧Dialog Robot Module

40‧‧‧Dialog robot control script

5‧‧‧Interactive oral learning materials

Claims (26)

A foreign language spoken learning system includes: a user module for controlling a user terminal operation, and the message output by the system is presented to a user in at least one of a voice and a text, and is responsible for Receiving a response voice sent by the user, and converting the response voice into a text input response; an interactive speaking learning teaching material, which is a plurality of courses designed for the interactive interaction between the system and the user, And supporting the system to provide at least one interactive learning mode; a system master module for controlling the operation of a server end, and leading and controlling interactive learning with the user, that is, according to the interaction Orsaid the content of a course in the learning material to output a message to the user, and compare whether the text input response converted by the response voice sent by the user meets a predetermined response corresponding to the message; And a user interface attached to the user module for presenting the content of the interactive oral speaking learning material and the system and the The interaction process between the user and the user; wherein the system master module has elasticity of the comparison of the user's response, so that the user's response can have a statement expression elasticity without changing the semantic meaning without Must be the same as the intended response. The foreign language spoken learning system according to claim 1, wherein the at least one interactive learning mode comprises a role dialogue mode, wherein the interactive voice is played by the system and the user respectively. Said to learn the two dialogue characters of a dialogue script in the teaching material for conducting a dialogue exercise; wherein the system will speak according to a statement of the dialogue script, and compare whether the response of the user is consistent with a reservation in the dialogue script The response statement, if not met, the system will ask the user to respond again, if it is met, the system will perform the statement of the dialog script. The foreign language spoken learning system as described in claim 1, wherein the at least one learning mode includes a request response mode, in which the system sends a request message to the user, and The user responds in a spoken manner, and then the system compares whether the response meets a predetermined response to the request message, and if the system does not comply, the user will respond to the user. The foreign language spoken learning system as described in claim 1, wherein the user interface further comprises a prompting instruction, and when the user module receives the prompting instruction, at least one prompt is provided. The foreign language spoken learning system as described in claim 4, wherein when the user module receives the prompting instruction, the user module converts a prompt text into a voice play. The foreign language spoken language learning system of claim 4, wherein when the user module receives the prompting instruction, the user module briefly flashes a prompt text on the user interface. For example, the foreign language spoken language learning system described in claim 1 is characterized in that any statement of the internal dialogue script of the interactive oral learning teaching material is composed of at least one synonym sentence. For example, in the foreign language spoken language learning system described in claim 7, wherein a speech of the system is selected from the synonymous sentence of the statement of the dialog script. The foreign language spoken learning system as described in claim 7 wherein, for a statement of the system, the correct response of the user may be any synonymous sentence of a predetermined response sentence in the dialog script. The foreign language spoken learning system as described in claim 1, wherein the at least one interactive learning mode further comprises a speaking test mode, wherein in the spoken test mode, the system will present a test question, and Passing through the user interface in at least one of voice and text The test topic is then answered by the user in a spoken manner, and the system determines whether the response meets a predetermined response to the test question. The foreign language spoken learning system as described in claim 1, wherein the interactive speaking learning teaching material comprises a dialogue script consisting of at least one synonym sentence, and the dialog script is presented in the user interface. The user interface includes a synonym sentence transformation instruction. When receiving the synonym sentence transformation instruction, the user interface performs a synonym replacement on the dialog script, that is, randomly changing at least one statement in the dialog script to the statement. Synonymous. For example, the foreign language spoken learning system described in claim 1 of the patent application, wherein the dialogue interaction between the system and the user and the comparison of the user responses are performed by the system control module by a dialog robot module In the process, the script control language of the dialog robot module will provide statement style matching flexibility, thereby supporting the system to respond to the user's response with the required flexibility. A foreign language speaking learning method is implemented by a foreign language speaking learning system capable of providing at least one learning mode, and the foreign language speaking learning system has a design designed for dialogue interaction characteristics with a user. An interactive oral learning learning material, and a user interface for presenting a course content of the interactive speaking teaching material and a dialogue interaction process between the system and the user, wherein the method comprises: by a system master The control module leads and controls the interactive learning between the system and the user, and sends a message to a user module according to the content of the course; the user module uses the message as a voice and text And outputting to the user at least one mode, and receiving a response voice from the user and converting the response voice into a text input response; Receiving the text input response by the system main control module, and comparing whether the text input response meets a predetermined response corresponding to the message; wherein the system main control module compares the user response It is flexible, so that the user's response can be expressed in the same semantic range without having to be the same as the predetermined response. The method for learning a foreign language spoken language as described in claim 13 further includes: by the system and the user respectively playing a dialogue role of a dialogue script in the interactive oral learning teaching material, and performing a dialogue exercise Providing a role dialogue mode of the at least one learning mode; wherein, in the character dialogue mode, the system performs a speech according to a statement of the dialog script, and compares whether the user's mouth response is consistent A predetermined response statement in the dialog script, if not met, the user is required to respond again, and if it is met, the statement of the dialog script is executed. The method for learning foreign language spoken words according to claim 13 of the patent application, further comprising: speaking by the system to present a request message to the user, and receiving a response from the user in a spoken manner, and The comparison determines whether the response meets a predetermined response of the request message to provide a required response mode of the learning mode; if not, the user is requested to respond again. The foreign language spoken learning method of claim 13, wherein the method further comprises: providing at least one prompt by the user module when the user module receives a prompt instruction through the user interface . The foreign language spoken learning method as described in claim 16 further includes: converting the prompt text into a voice play by the user module. For example, the foreign language oral learning method described in claim 16 of the patent application scope includes: A prompt text is briefly flashed on the user interface by the user module. For example, the foreign language spoken language learning method described in claim 13 of the patent application scope, wherein any statement of a dialogue script in the interactive oral learning learning textbook is composed of at least one synonym sentence. For example, the foreign language spoken learning method described in claim 19, wherein the system further comprises: selecting one of the synonymous sentences of a sentence in the dialog script as a speech. The foreign language spoken learning method as described in claim 19, further comprising: for a speech of the system, the correct response of the user may be any synonymous sentence of a predetermined response sentence in the dialog script. The foreign language spoken learning method as described in claim 13 , wherein the method further comprises: the system speaking to present a test question; and the user module presenting the test question in at least one of a voice and a text. And receiving a response from the user in a spoken manner; and the system compares a predetermined response to determine whether the response meets the test question to provide a one-stop test mode of the at least one interactive learning mode. The method for learning a foreign language spoken language according to claim 13 , wherein the method further comprises: presenting, by the user interface, a dialog script included in the interactive speaking teaching material, wherein any statement of the dialog script is And consisting of at least one synonym sentence; when the user module receives a synonym sentence transformation instruction, the user module replaces the conversation script on the user interface by synonymous sentence, and randomly selects the conversation script At least one statement is changed to a synonym for the statement. For example, the foreign language spoken learning method described in claim 13 of the patent application, wherein the dialogue interaction between the system and the user and the comparison of the user responses are borrowed through the system master module It is performed by a dialog robot module, whereby the script control language of the dialog robot module will provide statement style comparison flexibility, thereby supporting the system to respond to the user with the required flexibility. A computer program product, when loaded on an electronic device, will form the user module of the foreign language spoken learning system as described in claim 1. A computer program product, when loaded on a server, will form the system master module of the foreign language speaking learning system as described in claim 1.