阅读说明：本技术 一种基于人工智能和真人翻译的同步翻译系统及方法 (Synchronous translation system and method based on artificial intelligence and real person translation ) 是由 张运军 于 2018-08-01 设计创作，主要内容包括：本发明公开了一种基于人工智能和真人翻译的同步翻译系统及其方法,翻译系统包括用户终端,服务器,译员终端,专员终端四个模块。翻译方法包括用户及译员注册、用户发起翻译请求,生成订单、根据订单需求,选择机器翻译或人工翻译、根据需求匹配一定数量的译员供用户选择,确认译员、开始进行实时同步翻译、翻译结束,用户对翻译质量进行打分、根据用户打分结果,判断是否需要对译员进行升降级操作,并根据判断结果执行相应操作；将翻译内容转化为语料,通过机器学习提高机器翻译水平等七个步骤。本发明有助于完善远程在线翻译服务的译员分配系统,提高客户体验,促进在线翻译市场良性发展。(The invention discloses a synchronous translation system and a method thereof based on artificial intelligence and human translation. The translation method comprises the steps that a user and a translator register, the user initiates a translation request, an order is generated, machine translation or manual translation is selected according to the requirement of the order, a certain number of translators are matched according to the requirement for selection of the user, the translator is confirmed, real-time synchronous translation is started, translation is finished, the user scores the translation quality, whether the translator needs to be subjected to upgrading and downgrading operation or not is judged according to the scoring result of the user, and corresponding operation is executed according to the judgment result; and converting the translation content into a corpus, and improving the machine translation level through machine learning. The invention is beneficial to perfecting the translator distribution system of the remote online translation service, improving the customer experience and promoting the benign development of the online translation market.)

1. A synchronous translation system based on artificial intelligence and human translation is characterized by at least comprising a user terminal (10), a server (20) and a translator terminal (30);

the user terminal (10) comprises at least: a communication module (11) for communicating between internal modules of the translator terminal or communicating the translator terminal with the outside; a display module (12) for displaying input and output contents; a voice module (13) for inputting or outputting voice; a translation module (14) for implementing synchronous translation; a feedback module (15) for inputting feedback information of the user to the translator; the communication module (12) is electrically connected with the display module (11), the voice module (13) and the translation module (14);

the server (20) comprises at least: a communication module (21) for communication between the internal modules of the server or communication between the server and the outside; a data storage module (22) for storing information generated between the server (20) modules or external information received; a machine translation module (23) for implementing machine translation; the translator scoring module (24) is used for calculating the received information and returning a translator score value and grade; an order distribution module (25) for distributing the user's demand order to a particular translator;

said interpreter terminal (30) comprising at least: a communication module (31) for communicating between internal modules of the interpreter terminal or communicating the interpreter terminal with the outside; a display module (32) for displaying input and output contents; a voice module (33) for inputting or outputting voice; a translation module (34) for implementing synchronous translation; the communication module (31) is electrically connected with the display module (32), the voice module (33) and the translation module (34);

the user terminal (10), the server (20) and the translator terminal (30) communicate through a communication module (11), a communication module (21) and a communication module (31).

2. A synchronous translation system based on artificial intelligence and human translation according to claim 1, characterized in that said user terminal (10) further comprises a registration module (16) and a personal module (17);

the registration module (16) is used for receiving registration information input by a user, and the registration module (16) is electrically connected with the communication module (11) and the display module (12);

the personal module (17) is used for managing personal information of a user, and the personal module (17) is electrically connected with the communication module (11) and the display module (12).

3. The system of claim 1, wherein the server (20) further comprises a decision generation module (27) for generating decisions according to user requirements; the decision generation module (27) is electrically connected with the communication module (21) and the data storage module (22).

4. The system for synchronous translation based on artificial intelligence and human translation according to claim 1, wherein the translator terminal (30) further comprises a registration module (35) and a personal module (36);

the registration module (35) is used for receiving registration information input by a translator, and the registration module (35) is electrically connected with the communication module (31) and the display module (32);

the personal module (36) is used for managing personal information of the interpreter, and the personal module (36) is electrically connected with the communication module (31) and the display module (32).

5. The system for synchronous translation based on artificial intelligence and human translation as claimed in claim 1, further comprising a specialist terminal (40), wherein the specialist terminal (40) comprises at least: a communication module (41) for communicating between the internal modules of the specialist terminal or communicating between the translator terminal and the outside; a display module (42) for displaying input and output contents; the background processing module (43) is used for controlling the internal parameters of the system; the accent scoring module (44) is used for receiving the voice information of the interpreter and scoring the accent of the interpreter; the communication module (41) is electrically connected with the display module (42), the background processing module (43) and the accent scoring module (44);

the special member terminal (40), the user terminal (10), the server (20) and the translator terminal (30) communicate through a communication module (11), a communication module (21), a communication module (31) and a communication module (41).

6. The system for synchronous translation based on artificial intelligence and human translation according to claim 5, wherein said specialist terminal (40) further comprises a personal module (45) for managing personal information of an interpreter; the personal module (45) is electrically connected with the communication module (41) and the display module (42).

7. The system of claim 1, wherein the data storage module (22) comprises at least one of a temporary database (22a), a translator registration information database (22b), a user registration information database (22c), a translator scoring database (22d), a user feedback information database (22e), a decision information database (22f), a corpus learning database (22 g);

a temporary database (22a) for storing temporary information generated in the processing process of each module of the server (20);

an interpreter registration information database (22b) for storing registration information of an interpreter;

a user registration information database (22c) for storing registration information of users;

the translator scoring database (22d) is used for storing scoring and grade information of each field of the translator;

a user feedback information database (22e) for storing feedback information of the user to the translator;

a decision information database (22f) for storing various types of decision-related information required by the decision-making module (27);

and the corpus learning database (22g) is used for storing the corpus converted by the translation content of the translator.

8. The system for synchronous translation based on artificial intelligence and human translation according to claim 1, wherein the user terminal (10) further comprises a positioning module (18) for determining a geographical location of the user terminal (10); the positioning module (18) is electrically connected with the communication module (11) and the display module (12).

9. The system for synchronous translation based on artificial intelligence and human translation as claimed in claim 1, wherein said server further comprises a corpus learning module (26) for converting the translated content into corpus and increasing the machine translation level through machine learning; the communication module (21) is electrically connected with the data storage module (22), the machine translation module (23), the translator scoring module (24), the order distribution module (25) and the corpus learning module (26).

10. The system for synchronous translation based on artificial intelligence and human translation according to claim 1, wherein the server (20) can be disposed in the user terminal (10), or in the translator terminal (30), or in the specialist terminal (40), or the server (20) is independent from each terminal; the server (20) is a local server or a cloud server.

11. A synchronous translation method based on artificial intelligence and human translation is characterized by at least comprising the following steps:

step S002: a user initiates a translation request to generate an order;

the step S002 includes at least the following steps:

s002 a: verifying whether the user operates himself or herself, if the verification is successful, entering the step S002b, and if the verification is failed, stopping generating the order;

s002 b: the user selects whether to input a specific requirement, if the specific requirement is selected to be input, the step proceeds to step S002c, and if the specific requirement is not selected to be input, the step proceeds to step S002 d;

s002 c: the user selects to input at least one of the following:

user position, translation language, translation scene, whether manual translation is needed, required translator level, required translator adept field, instant translation or reserved translation, reserved translation time and default setting;

s002 d: automatically setting requirements according to the requirements of the user or the system, and generating an order;

step S003: selecting machine translation or manual translation according to order requirements, entering a step S005 if the machine translation is selected, and entering a step S004 if the manual translation is selected;

step S004: matching a plurality of translators according to requirements for selection by a user, and confirming the translators;

step S005: starting real-time synchronous translation, and translating the received language into another language;

step S006: after the translation is finished, scoring the translation quality by the user;

step S007: judging whether the translator needs to be subjected to upgrading and downgrading operation according to the user grading result, and executing corresponding operation according to the judgment result; and converting the translation content into a corpus, and improving the machine translation level through machine learning.

12. The synchronous translation method based on artificial intelligence and human translation of claim 11, wherein before step S002, the method comprises:

step S001: registering a user and an interpreter;

said step S001 comprises at least two parts,

s001 a: registering a user;

s001 b: the translator is registered.

13. The synchronous translation method based on artificial intelligence and human translation of claim 12, wherein in step S001, the user registers part S001a, and the content input by the user includes at least one of the following items:

the login information at least comprises one of a user name, a mobile phone number, a platform account and a login password;

the common service information at least comprises one of translation language, translation occasion, translation field, translation difficulty and default setting.

14. The synchronous translation method based on artificial intelligence and human translation of claim 12, wherein in step S001, the translator registering part S001b, the translator inputting the registration information comprises:

the identity authentication is used for authenticating the legal identity of the interpreter, and the certificate which is required to be provided by the interpreter at least comprises one of an identity card, a driving license and a passport;

basic information input at least comprising the name, the gender, the age, the translation language and the adequacy field of the translator;

inputting language and translation related information, and inputting an information certificate capable of proving the translation level of a translator;

auxiliary information input at least comprising one of overseas tourism, life experience, translation working time, qualification certificate or working experience of the adequacy field of the translator;

and the entry of the accent scoring at least comprises the step that the interpreter needs to finish the preset translation content within the specified time, and the accent scoring is scored by the accent evaluation software or professionals.

15. The synchronous translation method according to claim 12, wherein said step S001 of registering the translator in section S001b further comprises a subsequent step S001c of scoring and ranking the translator according to a predetermined scoring criteria based on the translator registration information.

16. The method for synchronous translation based on artificial intelligence and human translation according to claim 11, wherein the method for verifying whether the user operates himself/herself in step S002a includes at least one of the following methods:

fingerprint verification, password verification, short message verification, password card verification, slider verification, graph verification and verification code verification.

17. The method for synchronous translation based on artificial intelligence and human translation according to claim 11, wherein said step S003 at least comprises the following steps:

step S003 a: judging whether the user selects manual translation, if not, selecting machine translation, and executing the step S005; if the user selects manual translation, go to step S003 b:

step S003 b: judging whether the user selects the required translator level, and if the user selects the required translator level, executing the step S004; if the user does not select the interpreter level, go to step S003 c;

step S003 c: calculating and generating the most matched translator grade according to the specific requirement information of the user, and informing the grade information to the user;

step S003 d: the user needs to make a selection within a preset time range, whether the optimal matching grade generated by calculation is accepted or not is judged, and if the user selects to reject, the translator grade needs to be specified; if the user chooses to accept or no selection is made within a specified time, the interpreter level is determined to be the best match level generated by the calculation.

18. The synchronous translation method based on artificial intelligence and human translation of claim 11, wherein said step S004 comprises at least the following steps:

step S004 a: screening a plurality of available interpreters capable of receiving orders according to the user demand information and the confirmed interpreter grades;

step S004 b: sending the translator information obtained by screening to a user, and waiting for selection of the user;

step S004 c: and if the user selects overtime or abandons the selection, selecting one translator to be connected.

19. The method for synchronous translation based on artificial intelligence and human translation according to claim 18, wherein in step S004c, when the user selects timeout or abandons the selection, the method for selecting the translator includes at least one of the following two methods:

the first mode is as follows: randomly selecting a translator to be selected;

the second mode is as follows: and calculating to obtain a matched translator according to the demand information and the translator information.

20. The synchronous translation method based on artificial intelligence and human translation of claim 11, wherein the step S005 comprises at least the following steps:

step S005 a: detecting that a translation line is smooth, prompting a user to start translation, and waiting for confirmation of the user;

step S005 b: after the user confirms, real-time synchronous translation is started, and timing is started;

step S005 c: monitoring the line connection condition in the translation process, and stopping timing when translation is finished if the line is normal and smooth in the whole process; if the user and the translator contact each other in the translation process, go to step S005 d;

step S005 d: detecting a cause of the interruption and attempting reconnection; if the reconnection is successful, continuing the synchronous translation, and executing step S005 c; if it is detected that the reconnection fails due to the user reason, go to step S005 e; if the reconnection failure is detected due to the interpreter, go to step S005 f;

step S005 e: stopping timing, maintaining the translator connection state within a preset time, waiting for the user to connect, continuing the translation process if the user is successful in connecting within the preset time, and executing step S005 c; if the user does not successfully connect the line beyond the preset time, closing the communication of the translator end;

step S005 f: stopping timing, keeping the user connection state within a preset time, waiting for the translator to connect, continuing the translation process if the translator succeeds in connecting within the preset time, and executing step S005 c; if the translator is not successfully connected within the preset time, a matched translator meeting the conditions is connected immediately to carry out translation work;

the matched translators meeting the conditions refer to the translators which are screened out in the step S004 and can receive orders according to the user requirement information and the confirmed translator grades.

21. The synchronous translation method based on artificial intelligence and human translation according to claim 11, wherein the step S006 is executed with different results when:

in the first case: the line is normal in the translation process or the reconnection is successful after the line interruption, and the user scores according to the satisfaction degree, and then the scoring result of the user is recorded;

in the second case: if the line is normal in the translation process or the reconnection is successful after line interruption, the user gives up scoring, and the highest scoring in the preset scoring system is recorded by default;

in the third case: if the line is interrupted and the user connection fails in the translation process, the scoring result of the translation is not recorded;

in a fourth case: in the translation process, lines are interrupted and translators are replaced midway, then the user scores all translators participating in translation, and if the user scores the scores by himself, the scoring result of the user is recorded; if the user gives up the score, recording the highest score in the preset scoring system by default.

22. The method for synchronous translation based on artificial intelligence and human translation of claim 11, wherein the step S007 at least comprises the following steps:

step S007 a: acquiring an initial score and an initial grade of an interpreter;

step S007 b: correspondingly increasing or decreasing the grade of the translator according to the current grade of the user received by the translator;

step S007 c: if the grade of the interpreter is larger than the grade interval of the current grade, upgrading the interpreter, and if the grade of the interpreter is smaller than the grade interval of the current grade, degrading the interpreter.

23. The synchronous translation method based on artificial intelligence and human translation of claim 11, wherein the step S007 follows the following corpus learning rule:

and calculating the learning weight of the corpus formed by the current translation content according to the user score and the translator level, and performing machine learning according to the learning weight of different corpora.

24. The synchronous translation method based on artificial intelligence and human translation of claim 11, wherein the learning weight and the translator level are positively correlated to the user score in the corpus learning rule.

25. A synchronous translation system based on artificial intelligence and human translation, comprising a user terminal (10) and a server (20), characterized by comprising any one of the synchronous translation methods based on artificial intelligence and human translation of claims 11 to 23.

Technical Field

The invention relates to a language translation and evaluation technology, in particular to a synchronous translation system and method based on artificial intelligence and human translation.

Background

International import and export trade and communication of various technologies, cultures, tourism and the like present increasingly frequent situations, and different languages as a gap seriously obstruct smooth communication between people. At present, over 1000 ten thousand foreign language clients are received internationally in hotels, shopping malls, scenic spots and the like with foreign language communication demands, and the foreign language communication demands are long-term, high-repeatability and continuous demands! The current artificial intelligent speech translation cannot meet the communication requirement, and in addition, the artificial translation auxiliary communication is more needed in some fields with stronger speciality and commercial fields. Compared with machine translation, manual translation has good translation quality, but also faces high cost, and the cost is not only from the service cost of a translator, but also from the food and sleeping cost, the air ticket cost, the ground transportation cost and other related costs of the translator. In summary, the cost is so high that the manual translation is a service which is difficult to popularize for a long time in the past, and the ordinary consumers cannot enjoy the convenience brought by the manual translation.

The coming of the internet era and the rapid development of the communication field greatly reduce the cost of remote communication among people, so that a translator can remotely provide translation services, and translation services based on communication modes such as a telephone and a 4G network are in a rapidly increasing trend. Under the condition of different application scenes, the requirements of people on the translation accuracy and the translation effectiveness are different, for example, in some important meeting places, the translation content is generally complex, qualified simultaneous interpretation personnel are needed to ensure the translation quality and the real-time performance, the translation cost is also high and generally exceeds 1 ten thousand yuan per hour; in tourist attractions and environments such as hotels and shopping malls, only relatively simple communication is needed, the requirement on translation real-time performance is relatively low, simultaneous interpretation is not needed, translators at a primary level can be competent even machine translation, and the translation fee is low. Therefore, how to match translators or machine translation with different translation levels according to application scenes and customer requirements enables customers to spend more reasonably on the premise of meeting the translation requirements, can improve the customer experience level, is beneficial to the benign development of an online translation market, and can ensure the optimal allocation of translator resources under the market environment which is weak to demand.

Disclosure of Invention

The invention aims to provide a synchronous translation system and method based on artificial intelligence and real person translation, so that reasonable butt joint of translation service for users without level requirements is realized, translator resources are effectively distributed, user experience is improved, and benign development of an online translation market is promoted.

To achieve the above and other related objects, the present invention provides a synchronous translation system based on artificial intelligence and human translation, characterized by comprising at least a user terminal (10), a server (20), a translator terminal (30);

the user terminal (10) comprises at least: a communication module (11) for communicating between internal modules of the translator terminal or communicating the translator terminal with the outside; a display module (12) for displaying input and output contents; a voice module (13) for inputting or outputting voice; a translation module (14) for implementing synchronous translation; a feedback module (15) for inputting feedback information of the user to the translator; the communication module (12) is electrically connected with the display module (11), the voice module (13) and the translation module (14);

the server (20) comprises at least: a communication module (21) for communication between the internal modules of the server or communication between the server and the outside; a data storage module (22) for storing information generated between the server (20) modules or external information received; a machine translation module (23) for implementing machine translation; the translator scoring module (24) is used for calculating the received information and returning a translator score value and grade; an order distribution module (25) for distributing the user's demand order to a particular translator; the corpus learning module (26) is used for converting the translation content into corpus and improving the machine translation level through machine learning; the communication module (21) is electrically connected with the data storage module (22), the machine translation module (23), the translator scoring module (24), the order distribution module (25) and the corpus learning module (26);

the interpreter terminal (30) comprises at least: a communication module (31) for communicating between internal modules of the interpreter terminal or communicating the interpreter terminal with the outside; a display module (32) for displaying input and output contents; a voice module (33) for inputting or outputting voice; a translation module (34) for implementing synchronous translation; the communication module (31) is electrically connected with the display module (32), the voice module (33) and the translation module (34);

the user terminal (10), the server (20) and the translator terminal (30) communicate through a communication module (11), a communication module (21) and a communication module (31).

Preferably, the user terminal (10) further comprises a registration module (16) for receiving registration information input by the user; the registration module (16) is electrically connected with the communication module (11) and the display module (12).

Preferably, the user terminal (10) further comprises a personal module (17) for managing personal information of the user; the personal module (17) is electrically connected to the communication module (11) and the display module (12).

Preferably, the server (20) further comprises a decision generating module (27) for generating a decision according to the user's needs; the decision generation module (27) is electrically connected with the communication module (21) and the data storage module (22).

Preferably, the translator terminal (30) further comprises a registration module (35) for receiving registration information input by the translator; the registration module (35) is electrically connected to the communication module (31) and the display module (32).

Preferably, the translator terminal (30) further comprises a personal module (36) for managing personal information of the translator; the personal module (36) is electrically connected to the communication module (31) and the display module (32).

The synchronous translation system of artificial intelligence and real person translation still includes special person terminal (40), and special person terminal (40) includes at least: a communication module (41) for communicating between the internal modules of the specialist terminal or communicating between the translator terminal and the outside; a display module (42) for displaying input and output contents; the background processing module (43) is used for controlling the internal parameters of the system; the accent scoring module (44) is used for receiving the voice information of the interpreter and scoring the accent of the interpreter; the communication module (41) is electrically connected with the display module (42), the background processing module (43) and the accent scoring module (44);

the special member terminal (40), the user terminal (10), the server (20) and the translator terminal (30) communicate through a communication module (11), a communication module (21), a communication module (31) and a communication module (41).

Preferably, the specialist terminal (40) further comprises a personal module (45) for managing personal information of the interpreter; the personal module (45) is electrically connected to the communication module (41) and the display module (42).

Preferably, the data storage module (22) comprises at least one of a temporary database (22a), a translator registration information database (22b), a user registration information database (22c), a translator scoring database (22d), a user feedback information database (22e), a decision information database (22f), and a corpus learning database (22 g);

a temporary database (22a) for storing temporary information generated in the processing process of each module of the server (20);

an interpreter registration information database (22b) for storing registration information of an interpreter;

a user registration information database (22c) for storing registration information of users;

the translator scoring database (22d) is used for storing scoring and grade information of each field of the translator;

a user feedback information database (22e) for storing feedback information of the user to the translator;

a decision information database (22f) for storing various types of decision-related information required by the decision-making module (27);

a corpus learning database (22g) for storing a corpus into which the translation contents of the translator are converted;

preferably, the user terminal (10) further comprises a positioning module (18) for determining the geographical location of the user terminal (10); the positioning module (18) is electrically connected with the communication module (11) and the display module (12).

Preferably, the server (20) can be arranged in the user member terminal (10), or in the translator terminal (30), or in the specialist terminal (40), or the server (20) is independent of each terminal; the server (20) is a local server or a cloud server.

The invention also provides a synchronous translation method based on artificial intelligence and human translation, which is characterized by at least comprising the following steps:

step S002: a user initiates a translation request to generate an order;

step S002 includes at least the following steps:

s002 a: verifying whether the user operates himself or herself, if the verification is successful, entering the step S002b, and if the verification is failed, stopping generating the order;

s002 b: the user selects whether to input a specific requirement, if the specific requirement is selected to be input, the step proceeds to step S002c, and if the specific requirement is not selected to be input, the step proceeds to step S002 d;

s002 c: the user selects to input at least one of the following:

user position, translation language, translation scene, whether manual translation is needed, required translator level, required translator adept field, instant translation or reserved translation, reserved translation time and whether default setting is selected;

s002 d: automatically setting requirements according to the requirements of the user or the system, and generating an order;

step S003: selecting machine translation or manual translation according to order requirements, entering a step S005 if the machine translation is selected, and entering a step S004 if the manual translation is selected;

step S004: matching a plurality of translators according to requirements for selection by a user, and confirming the translators;

step S005: starting real-time synchronous translation, and translating the received language into another language;

step S006: after the translation is finished, scoring the translation quality by the user;

step S007: judging whether the translator needs to be subjected to upgrading and downgrading operation according to the user grading result, and executing corresponding operation according to the judgment result; and converting the translation content into a corpus, and improving the machine translation level through machine learning.

Preferably, step S002 is preceded by:

step S001: registering a user and an interpreter;

the step S001 comprises at least two parts,

s001 a: registering a user;

s001 b: the translator is registered.

Preferably, in step S001, the user registration part S001a, the content to be input by the user includes at least one of the following:

the login information at least comprises one of a user name, a mobile phone number, a platform account and a login password;

the common service information at least comprises one of translation language, translation occasion, translation field, translation difficulty and default setting.

Preferably, the step S001 of the interpreter registering section S001b, the interpreter needs to input the registration information including:

the identity authentication is used for authenticating the legal identity of the interpreter, and the certificate which is required to be provided by the interpreter at least comprises one of an identity card, a driving license and a passport;

basic information input at least comprising the name, the gender, the age, the translation language and the adequacy field of the translator;

inputting language and translation related information, and inputting an information certificate capable of proving the translation level of a translator;

auxiliary information input at least comprising one of overseas tourism, life experience, translation working time, qualification certificate or working experience of the adequacy field of the translator;

and the entry of the accent scoring at least comprises the step that the interpreter needs to finish the preset translation content within the specified time, and the accent scoring is scored by the accent evaluation software or professionals.

Preferably, the translator registering part S001b of the step S001 further includes a subsequent step S001c of scoring and ranking the translators according to a preset scoring criterion based on the translator registration information.

Preferably, the method for verifying whether the user operates himself or herself in step S002a includes at least one of the following methods:

fingerprint verification, password verification, short message verification, password card verification, slider verification, graph verification and verification code verification.

Preferably, step S003 includes at least the steps of:

step S003 a: judging whether the user selects manual translation, if not, selecting machine translation, and executing the step S005; if the user selects manual translation, go to step S003 b:

step S003 b: judging whether the user selects the required translator level, and if the user selects the required translator level, executing the step S004; if the user does not select the interpreter level, go to step S003 c;

step S003 c: calculating and generating the most matched translator grade according to the specific requirement information of the user, and informing the grade information to the user;

step S003 d: the user needs to make a selection within a preset time range, whether the optimal matching grade generated by calculation is accepted or not is judged, and if the user selects to reject, the translator grade needs to be specified; if the user chooses to accept or no selection is made within a specified time, the interpreter level is determined to be the best match level generated by the calculation.

Preferably, step S004 includes at least the following steps:

step S004 a: screening a plurality of available interpreters capable of receiving orders according to the user demand information and the confirmed interpreter grades;

step S004 b: sending the translator information obtained by screening to a user for selection;

step S004 c: connecting translators selected by the user, and if the user selects overtime or abandons the selection, selecting one translator for connection;

preferably, in step S004c, when the user selects time-out or abandons the selection, the method of selecting the interpreter includes at least one of the following two ways:

the first mode is as follows: randomly selecting a translator to be selected;

the second mode is as follows: and calculating to obtain a matched translator according to the demand information and the translator information.

Preferably, the matching translator is the best translator.

Preferably, step S005 comprises at least the steps of:

step S005 a: detecting that a translation line is smooth, prompting a user to start translation, and waiting for confirmation of the user;

step S005 b: after the user confirms, real-time synchronous translation is started, and timing is started;

step S005 c: monitoring the line connection condition in the translation process, and stopping timing when translation is finished if the line is normal and smooth in the whole process; if the user and the translator contact each other in the translation process, go to step S005 d;

step S005 d: detecting a cause of the interruption and attempting reconnection; if the reconnection is successful, continuing the synchronous translation, and executing step S005 c; if it is detected that the reconnection fails due to the user reason, go to step S005 e; if the reconnection failure is detected due to the interpreter, go to step S005 f;

step S005 e: the timing is stopped, the translator is kept connected for a predetermined time, the user is waited for connection, if the user is successful in connecting within the predetermined time, the translation process is continued, and step S005c is executed. If the user does not successfully connect the line beyond the preset time, closing the communication of the translator end;

step S005 f: the timing is stopped, the user connection status is maintained for a predetermined time, the translator is waited for connection, if the translator is successful in the connection within the predetermined time, the translation process is continued, and step S005c is executed. If the translator is not successfully connected within the preset time, a matched translator meeting the conditions is connected immediately to carry out translation work;

the matched translators meeting the conditions refer to the translators which are screened out in the step S004 and can receive orders according to the user requirement information and the confirmed translator grades.

Preferably, step S006 is executed with different results when:

in the first case: the line is normal in the translation process or the reconnection is successful after the line interruption, and the user scores according to the satisfaction degree, and then the scoring result of the user is recorded;

in the second case: if the line is normal in the translation process or the reconnection is successful after line interruption, the user gives up scoring, and the highest scoring in the preset scoring system is recorded by default;

in the third case: if the line is interrupted and the user connection fails in the translation process, the scoring result of the translation is not recorded;

in a fourth case: in the translation process, lines are interrupted and translators are replaced midway, then the user scores all translators participating in translation, and if the user scores the scores by himself, the scoring result of the user is recorded; if the user gives up the score, recording the highest score in the preset scoring system by default;

preferably, step S007 includes at least the following steps:

step S007 a: acquiring an initial score and an initial grade of an interpreter;

step S007 b: correspondingly increasing or decreasing the grade of the translator according to the current grade of the user received by the translator;

step S007 c: if the grade of the interpreter is larger than the grade interval of the current grade, upgrading the interpreter, and if the grade of the interpreter is smaller than the grade interval of the current grade, degrading the interpreter.

Preferably, step S007 follows the corpus learning rules of:

and calculating the learning weight of the corpus formed by the current translation content according to the user score and the translator level, and performing machine learning according to the learning weight of different corpora.

Preferably, in the corpus learning rule, the learning weight and the interpreter level are positively correlated with the user score.

The invention also provides a synchronous translation system based on artificial intelligence and human translation, which comprises a user terminal (10) and a server (20) and is characterized by comprising any one of the synchronous translation methods based on artificial intelligence and human translation.

Specifically, the user terminal (10) of the interpreter scoring system at least comprises a processor (CPU), a memory (such as a hard disk, an SSD solid state disk, a DDR3, or a DDR4 memory), a camera, a communication chip (such as a Wifi module, a 4G module), and the like, and the chips may be integrated on one motherboard or may be integrated on multiple motherboards respectively. A Graphics Processing Unit (GPU) voice recognition chip, a sound card (voice input output chip), a microphone may also be included. More specifically, the device can be a current mobile phone, a computer, an intelligent wearable device, a game device and the like.

The server (20) may include a processor (CPU), a Graphics Processing Unit (GPU), a memory (e.g., a hard disk, an SSD solid state disk, a DDR3 or a DDR4 memory, etc.), a voice recognition chip, a sound card (a voice input/output chip), a text-to-speech conversion chip, a communication chip (e.g., a Wifi, 4G module), etc., which may be integrated on one motherboard or may be integrated on multiple motherboards respectively. Of course, the server may include only some of the chips, as long as the functions are satisfied.

The interpreter terminal (30) comprises a processor (CPU), a Graphic Processing Unit (GPU), a memory (such as a hard disk, an SSD (solid state disk), a DDR3 or DDR4 memory and the like), a voice recognition chip, a sound card (a voice input and output chip), a text voice conversion chip, a communication chip (such as a Wifi (wireless fidelity), a 4G (global) module) and the like, wherein the chips can be integrated on one mainboard or a plurality of mainboards respectively.

The specialist terminal (40) comprises a processor (CPU), a Graphic Processing Unit (GPU), a memory (such as a hard disk, an SSD (solid state disk), a DDR3 or DDR4 memory and the like), a voice recognition chip, a sound card (a voice input and output chip), a text voice conversion chip, a communication chip (such as Wifi, a 4G module and a 5G module) and the like, wherein the chips can be integrated on one mainboard and can also be respectively integrated on a plurality of mainboards.

As above, the synchronous translation system and method based on artificial intelligence and human translation of the invention have the following beneficial effects:

(1) according to the invention, translators are classified into classes and fields, so that detailed translator matching requirements can be met, and translators are reasonably distributed to the maximum extent;

(2) the invention scores the translators from a plurality of angles such as translation qualification, working time, accent and the like, thereby realizing accurate and objective evaluation of the translation level of the translators;

(3) according to the translator grade change rule provided by the invention, the user feedback information is acted on the initial scoring of the translator, so that the initial scoring result is corrected, the translator can be promoted to consciously improve the translation level of the translator, and the translator with poor translation quality can be eliminated in time;

(4) the verification operation before the translation is started can effectively prevent economic loss caused by misoperation of a user or embezzlement of the user terminal by other people;

(5) the synchronous translation system and method based on artificial intelligence and real person translation can automatically evaluate the grade of a required translator according to the actual requirement of a user, match the best translator, and reduce the unnecessary cost of the user on the premise of ensuring the translation quality;

(6) the synchronous translation system and method based on artificial intelligence and real person translation can quickly match and connect a new optimal translator under the condition that a human translation translator loses connection and reconnection fails, and ensure that the translation process of a user is not influenced by network interruption in the aspect of the translator;

(7) according to the corpus learning rule provided by the invention, different learning weights are added to the corpus according to the translator grade and the user feedback result, so that the weights of the learning corpus converted from excellent translators and widely-evaluated translation contents in a training set can be effectively improved, the learning quality of machine learning is further improved, and the level of machine translation is rapidly improved.

Drawings

FIG. 1 is a schematic diagram of a system for providing a synchronous translation system based on artificial intelligence and human translation for the present invention

FIG. 2 is a schematic diagram showing components of a data storage module in the synchronous translation system based on artificial intelligence and human translation provided by the invention. (ii) a

FIG. 3 is a flow chart of a method for synchronous translation based on artificial intelligence and human translation proposed by the present invention;

FIG. 4 is a table showing scoring weight ratios statistics for different registration information of an interpreter according to the present invention;

FIG. 5 is a table showing scoring ranges and service category statistics for different translators according to the present invention;

FIG. 6 is a flowchart illustrating an embodiment of step S002 of the synchronous translation method according to the present invention;

FIG. 7 is a flowchart illustrating a specific implementation of step S003 of the synchronous translation method proposed by the present invention;

FIG. 8 is a flowchart illustrating a specific implementation process of step S004 of the synchronous translation method proposed by the present invention;

FIG. 9 is a flowchart illustrating an embodiment of step S005 of the synchronous translation method according to the present invention;

FIG. 10 is a flowchart illustrating an exemplary implementation of the translator upgrade portion of step S007 of the synchronous translation method of the present invention;

Detailed Description

The following description of the embodiments of the present invention is provided for illustrative purposes, and other advantages and effects of the present invention will become apparent to those skilled in the art from the present disclosure.

It should be understood that the structures, ratios, sizes, and the like shown in the drawings and described in the specification are only used for matching with the disclosure of the specification, so as to be understood and read by those skilled in the art, and are not used to limit the conditions under which the present invention can be implemented, so that the present invention has no technical significance, and any structural modification, ratio relationship change, or size adjustment should still fall within the scope of the present invention without affecting the efficacy and the achievable purpose of the present invention. In addition, the terms "upper", "lower", "left", "right", "middle" and "one" used in the present specification are for clarity of description, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not to be construed as a scope of the present invention.