CN111008533B - Method, device, equipment and storage medium for obtaining translation model - Google Patents
Method, device, equipment and storage medium for obtaining translation model Download PDFInfo
- Publication number
- CN111008533B CN111008533B CN201911251122.7A CN201911251122A CN111008533B CN 111008533 B CN111008533 B CN 111008533B CN 201911251122 A CN201911251122 A CN 201911251122A CN 111008533 B CN111008533 B CN 111008533B
- Authority
- CN
- China
- Prior art keywords
- translation model
- corpus
- target
- trained
- parallel corpus
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Abstract
The embodiment of the disclosure discloses a method, a device, equipment and a storage medium for acquiring a translation model, wherein the method comprises the following steps: acquiring a parallel corpus pair set, and respectively acquiring a first general translation model and a second general translation model after training through the parallel corpus pair set; acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and further forming a first pseudo parallel corpus pair set; and directionally training the trained first general translation model through the first pseudo parallel corpus pair set to obtain a first target translation model. According to the technical scheme of the embodiment of the disclosure, the corresponding language translation model can still be established under the condition that the parallel corpus pair in the specific field is lacked, and the accuracy of corpus translation in the specific field is greatly improved.
Description
Technical Field
The embodiment of the disclosure relates to software technologies, and in particular, to a method, an apparatus, a device, and a storage medium for acquiring a translation model.
Background
With the continuous development of computer technology, various translation software appears in the visual field of people, and becomes an important channel for people to acquire external information.
The existing translation software, the language translation model of which is usually built based on continuous training of a large number of monolingual parallel corpora (for example, monolingual parallel corpora composed of chinese documents and corresponding english documents) is used to implement directional translation (for example, chinese translation), but it is not easy to obtain a large number of parallel corpora, and especially, it is very difficult to obtain parallel corpora related to a specific field (for example, chinese medicine), so that the built language translation model is very poor in accuracy under the condition that a large number of parallel corpora cannot be obtained.
Disclosure of Invention
The disclosure provides a translation model acquisition method, a translation model acquisition device, translation model acquisition equipment and a translation model storage medium, so that a language translation model in a specific field is established under the condition that parallel corpus pairs in the field are lacked.
In a first aspect, an embodiment of the present disclosure provides a method for obtaining a translation model, including:
acquiring a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora;
respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set to obtain a trained first general translation model and a trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model;
acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo parallel corpus pair set;
and directionally training the trained first general translation model through the first pseudo parallel corpus pair set to obtain a first target translation model.
In a second aspect, an embodiment of the present disclosure provides an apparatus for obtaining a translation model, including:
the parallel corpus pair set acquisition module is used for acquiring a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora;
the initial training execution module is used for respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set so as to obtain the trained first general translation model and the trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model;
the first pseudo parallel corpus pair acquisition module is used for acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo parallel corpus pair set;
and the first target translation model acquisition module is used for directionally training the trained first general translation model through the first pseudo parallel corpus pair set to acquire a first target translation model.
In a third aspect, an embodiment of the present disclosure provides an electronic device, which includes a memory, a processing apparatus, and a computer program stored in the memory and executable on the processing apparatus, where the processing apparatus implements a method for obtaining a translation model according to any embodiment of the present disclosure when executing the computer program.
In a fourth aspect, embodiments of the present disclosure provide a storage medium containing computer-executable instructions, which when executed by a computer processor, are configured to perform the method for obtaining a translation model according to any of the embodiments of the present disclosure.
According to the technical scheme of the embodiment of the disclosure, the parallel corpus pair set is obtained, the two general translation models are trained respectively to obtain the trained general translation models respectively, the source language corpus set in the specific field forms the pseudo parallel corpus pair set by using one general translation model, and then the translation model in the target field is obtained according to the pseudo parallel corpus pair set and the other general translation model, so that the corresponding language translation model can be still established under the condition that the parallel corpus pair in the specific field is lacked, and the corpus translation accuracy in the specific field is greatly improved.
Drawings
The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent by referring to the following detailed description when taken in conjunction with the accompanying drawings. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and features are not necessarily drawn to scale.
Fig. 1 is a flowchart of a method for obtaining a translation model in a first embodiment of the present disclosure;
fig. 2 is a block diagram of a translation model obtaining apparatus in a second embodiment of the disclosure;
fig. 3 is a block diagram of a device in a third embodiment of the present disclosure.
Detailed Description
Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it is to be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein, but rather are provided for a more thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the disclosure are for illustration purposes only and are not intended to limit the scope of the disclosure.
It should be understood that the various steps recited in the method embodiments of the present disclosure may be performed in a different order, and/or performed in parallel. Moreover, method embodiments may include additional steps and/or omit performing the illustrated steps. The scope of the present disclosure is not limited in this respect.
The term "include" and variations thereof as used herein are open-ended, i.e., "including but not limited to". The term "based on" is "based, at least in part, on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Relevant definitions for other terms will be given in the following description.
It should be noted that the terms "first", "second", and the like in the present disclosure are only used for distinguishing different devices, modules or units, and are not used for limiting the order or interdependence relationship of the functions performed by the devices, modules or units.
It is noted that references to "a", "an", and "the" modifications in this disclosure are intended to be illustrative rather than limiting, and that those skilled in the art will recognize that "one or more" may be used unless the context clearly dictates otherwise.
The names of messages or information exchanged between devices in the embodiments of the present disclosure are for illustrative purposes only, and are not intended to limit the scope of the messages or information.
Example one
Fig. 1 is a flowchart of a method for obtaining a translation model according to an embodiment of the present disclosure, where the embodiment is applicable to a case where a language translation model of a specific domain is built in the absence of a parallel corpus pair of the specific domain, and the method may be executed by a device for obtaining a translation model according to an embodiment of the present disclosure, where the device may be implemented by software and/or hardware and integrated in an application program, and the method specifically includes the following steps:
s110, acquiring a parallel corpus pair set; the set of parallel corpus pairs comprises at least one parallel corpus pair, and each parallel corpus pair comprises paired source language corpus and target language corpus.
The parallel corpus pair is a corresponding corpus between two languages, including a source language corpus and a target language corpus, the source language is a language capable of guiding out another language, and the guided language is a target language, for example, the Chinese-English parallel corpus pair includes a Chinese document and a corresponding English document, if a translation model is used for performing Chinese-English translation operation, the Chinese document is the source language corpus, and the English document is the target language corpus; the parallel corpus pairs included in the set of parallel corpus pairs are all parallel corpus pairs of the same language type, that is, the source language between each parallel corpus pair is the same, and the target language is also the same, for example, the parallel corpus pairs included in the set of parallel corpus pairs are all chinese-english parallel corpus pairs.
Optionally, in this disclosure, the parallel corpus pair includes a text parallel corpus pair or a speech parallel corpus pair, that is, the parallel corpus pair may be a text material or a speech material.
S120, respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set to obtain a trained first general translation model and a trained second general translation model; and the source language corpus of the first general translation model is the target language corpus of the second general translation model, and the target language corpus of the first general translation model is the source language corpus of the second general translation model.
Optionally, in this disclosure, the first general translation model and the second general translation model may each include a sequence-to-sequence model; sequence to Sequence (seq 2seq) model, which is a neural network of Encoder-Decoder structure, the input is a Sequence (Sequence) and the output is also a Sequence; in the Encoder, a variable-length sequence is converted into a fixed-length vector expression, and the Decoder converts the fixed-length vector into a variable-length target signal sequence, so as to realize the input of an indefinite length to an indefinite-length output. The sequence-to-sequence model may include various types, for example, a seq2seq model based on a Recurrent Neural Network (RNN), a seq2seq model based on a Convolution Operation (CONV), and the like; the first and second generic translation models may further include a Right-to-Left (RTL) model; in the embodiment of the present disclosure, optionally, the types of the first general translation model and the second general translation model are not particularly limited.
The two parallel corpora included in the parallel corpus pair are respectively used as source language corpora of the first general translation model and the second general translation model, and then the corresponding other parallel corpora is respectively used as target language corpora, for example, by taking the Chinese-English parallel corpus pair in the above technical scheme as an example, the English parallel corpus is used as the source language corpora of the first general translation model, namely, the input side; taking the Chinese parallel corpus as a target language corpus of the first universal translation model, namely an output side; on the contrary, the Chinese parallel language material is used as the source language material of the second universal translation model; taking the English parallel language material as a target language material of the second universal translation model; the first general translation model after training is used as an English translation middle translation model, and the second general translation model after training is used as a middle translation English translation model.
S130, obtaining a first source language corpus set of a specific field, obtaining a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming a first pseudo parallel corpus pair set by the first source language corpus set of the specific field and the matched first target language corpus set.
For some specific fields, there is often only a single language corpus and there is no corresponding parallel corpus, for example, in the field of chinese medicine, a large amount of chinese corpora and also english corpora can be obtained, but parallel corpora corresponding to each other in chinese and english are difficult to obtain. Therefore, after a first source language corpus set of a specific field is obtained, the corpus set is input into a trained second universal translation model, and a first target language corpus set matched with the first source language corpus set of the specific field is obtained; the trained second universal translation model is used for learning vocabulary association and a grammatical structure and has a translation function, so that the acquired first target language corpus set and the first source language corpus set in the specific field can form a parallel corpus pair set, and the parallel corpus pair in the parallel corpus pair set is not a standard corpus pair but is acquired through the second universal translation model, so that the set is marked as a pseudo parallel corpus pair, namely a first pseudo parallel corpus pair, and the corresponding pseudo parallel corpus pair set is a first pseudo parallel corpus pair set; taking the above technical solution as an example, taking the acquired chinese corpus set in the field of traditional Chinese medicine as an input of a chinese-to-english translation model (i.e. a trained second universal translation model), so as to acquire a corresponding english corpus set, and form a first pseudo-parallel corpus pair set; alternatively, in the embodiments of the present disclosure, the type of a field in a specific field is not particularly limited.
S140, directionally training the trained first general translation model through the first pseudo parallel corpus pair set to obtain a first target translation model.
By taking the above technical scheme as an example, the translation model in the english translation (i.e. the trained first general translation model) is directionally trained through the first pseudo parallel corpus pair set in the field of traditional Chinese medicine, and accordingly, the translation model in the english translation in the field of traditional Chinese medicine can still be obtained without the chinese-english parallel corpus pair in the field of traditional Chinese medicine.
Optionally, in this embodiment of the present disclosure, performing directional training on the trained first general translation model through the first pseudo-parallel corpus pair set to obtain a first target translation model, including: and directionally training the trained first general translation model through the first pseudo parallel corpus pair set and the parallel corpus pair set to obtain a first target translation model. When the trained first general translation model is directionally trained, the first pseudo parallel corpus pair set and the parallel corpus pair set can be trained together on the first general translation model, so that the first general translation model can not lose training effects in other fields due to the introduction of the corpus in the specific field.
Optionally, in this embodiment of the present disclosure, after obtaining the first target translation model, the method further includes; judging whether the first target translation model meets the convergence requirement or not; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample; if the first target translation model does not meet the convergence requirement, the first target translation model is trained continuously through the first pseudo parallel corpus pair set until the first target translation model converges. Detecting the first target translation model by using a detection sample, matching the translation result of the detection sample with the real parallel corpus of the detection sample, and if the matching degree exceeds a certain preset threshold value, determining that the first target translation model meets the convergence requirement; the acquired first target translation model can be sent to a third-party detection platform and is detected by the third-party detection platform, if a valid instruction returned by the third-party detection platform is acquired, the first target translation model is judged to be valid, if an invalid instruction returned by the third-party detection platform is acquired, the first target translation model is judged to be invalid, and training is continued until the translation model passes the detection of the third-party detection platform.
Optionally, in this embodiment of the present disclosure, after the initial training is performed on the first general translation model and the second general translation model respectively through the set of parallel corpus pairs to obtain the trained first general translation model and the trained second general translation model, the method further includes: acquiring a second source language corpus set of a specific field, acquiring a second target language corpus set matched with the second source language corpus set of the specific field through the trained first universal translation model, and forming a second pseudo parallel corpus pair set by the second source language corpus set of the specific field and the matched second target language corpus set; and directionally training the trained second universal translation model through the second pseudo parallel corpus pair set to obtain a second target translation model. Taking the above technical solution as an example, taking the obtained english corpus set in the field of traditional Chinese medicine as an input of a translation model in the english translation (i.e. the trained first general translation model), so as to obtain a corresponding chinese corpus set, and form a second pseudo-parallel corpus pair set; and (3) directionally training the middle-to-English translation model (namely the trained first general translation model) through the second pseudo parallel corpus pair set, so that the middle-to-English translation model in the field of traditional Chinese medicine can still be obtained under the condition that the middle-to-English parallel corpus pair in the field of traditional Chinese medicine does not exist. Correspondingly, after the second target translation model is obtained, the method also comprises the following steps; judging whether the second target translation model meets the convergence requirement or not; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample; and if the second target translation model does not meet the convergence requirement, continuing training the second target translation model through the second pseudo parallel corpus pair set until the second target translation model converges. Specifically, the first source language is a source language of the trained first universal translation model and corresponds to a target language of the trained second universal translation model, and the second source language is a source language of the trained second universal translation model and corresponds to a target language of the trained first universal translation model. In the above technical solution, pseudo parallel corpus pairs suitable for training of the other party are respectively constructed through interaction between the chinese-to-english translation model and the english-to-english translation model, and in the absence of the chinese-to-english parallel corpus pairs, the chinese-to-english translation model and the english-to-english translation model suitable for the field of traditional Chinese medicine are still constructed only through the chinese corpus and the english corpus in the field of traditional Chinese medicine.
Optionally, in this embodiment of the present disclosure, if a translation model in a general field is desired to be obtained, for example, a translation model common to news information classes (without distinguishing specific fields), the first source language corpus set and the second source language corpus set of the corresponding news information classes may be respectively input into the second general translation model after initial training and the first general translation model after initial training, and through the construction process in the foregoing technical solution, a target translation model suitable for the general translation field is constructed, for example, a third target translation model (an english translation model in an english translation) and a fourth target translation model (an intermediate translation model in an english translation) suitable for the general translation field.
Optionally, in this embodiment of the present disclosure, after the combining the first source language corpus set of the specific field and the matched first target language corpus set into the first pseudo-parallel corpus pair set, the method further includes: performing noise adding operation on the first pseudo parallel corpus pair set to generate a first pseudo parallel corpus pair set containing noise; the noise addition operation includes adding words, deleting words, and/or disordering the order of words in the corpus. Specifically, noise adding operation is performed on the source language corpus in the set by the first pseudo-parallel corpus, and the target language corpus is not changed, for example, the source language corpus is Chinese 'I Love life', after noise is added, the source language corpus is modified into 'I Love life', the target language corpus is English 'I Love Live' and is not modified, the set is trained on the second universal translation model by the first pseudo-parallel corpus to which the noise is added, so that the anti-noise (namely anti-interference) capability of the second universal translation model is increased, and the error correction capability is improved.
Optionally, in this embodiment of the present disclosure, after the combining the first source language corpus set of the specific field and the matched first target language corpus set into the first pseudo-parallel corpus pair set, the method further includes: and respectively performing word order reversal operation on a first source language corpus set and a first target language corpus set in the first pseudo-parallel corpus pair set to obtain a new first pseudo-parallel corpus pair set. For example, the source language corpus is Chinese 'Apple and Pear', is modified into 'Pear and Apple' after the word sequence is reversed, the target language corpus is English 'Apple and Pear', and is modified into 'Pear and Apple' after the word sequence is reversed, so that the parallel corpus pairs are more diverse, and the second general translation model is trained through the first pseudo parallel corpus pair set after the word sequence is reversed, so that the second general translation model strengthens the vocabulary association and the grammatical structure of directional translation.
Optionally, in this embodiment of the present disclosure, a plurality of second general translation models may also be initially trained through the parallel corpus pair set, so as to obtain a plurality of trained second general translation models; wherein each of the second generic translation models is of a different type; further acquiring a first source language corpus set of a specific field, acquiring a plurality of matched first target language corpus sets through each trained second universal translation model, and forming the first source language corpus set of the specific field and each matched first target language corpus set into a plurality of first pseudo-parallel corpus pair sets; and finally, directionally training the trained first general translation model through each first pseudo parallel corpus pair set to obtain a first target translation model. For example, the seq2seq model based on RNN, the seq2seq model based on CONV and the right-to-left model in the above technical solution are initially trained respectively through a parallel corpus pair set to obtain trained second general translation models respectively; respectively acquiring a matched English corpus set from an acquired Chinese corpus set in the field of traditional Chinese medicine through a trained seq2seq model based on RNN, a seq2seq model based on CONV and a right-left model, and respectively forming three first pseudo-parallel corpus pair sets with the Chinese corpus set; and directionally training the trained first general translation model by using the three first pseudo-parallel corpus pair sets to obtain a first target translation model (namely a translation model in the English translation in the field of traditional Chinese medicine). In view of the above, through the monolingual speech material set in the field of traditional Chinese medicine, a plurality of pseudo parallel speech material pair sets are formed, the diversity of parallel speech material pairs in the field of traditional Chinese medicine is increased, and the noise immunity of the first target translation model is improved.
According to the technical scheme of the embodiment of the disclosure, the parallel corpus pair set is obtained, the two general translation models are trained respectively to obtain the trained general translation models respectively, the source language corpus set in the specific field forms the pseudo parallel corpus pair set by using one general translation model, and then the translation model in the target field is obtained according to the pseudo parallel corpus pair set and the other general translation model, so that the corresponding language translation model can be still established under the condition that the parallel corpus pair in the specific field is lacked, and the corpus translation accuracy in the specific field is greatly improved.
Example two
Fig. 2 is a block diagram of a structure of an apparatus for obtaining a translation model according to a second embodiment of the present disclosure, which specifically includes: a parallel corpus pair set obtaining module 210, an initial training executing module 220, a first pseudo-parallel corpus pair obtaining module 230, and a first target translation model obtaining module 240.
A parallel corpus pair set obtaining module 210, configured to obtain a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora;
an initial training execution module 220, configured to perform initial training on the first general translation model and the second general translation model respectively through the parallel corpus pair set, so as to obtain a trained first general translation model and a trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model;
a first pseudo-parallel corpus pair obtaining module 230, configured to obtain a first source language corpus set of a specific field, obtain a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and combine the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo-parallel corpus pair set;
a first target translation model obtaining module 240, configured to perform directional training on the trained first general translation model through the first pseudo-parallel corpus pair set, so as to obtain a first target translation model.
According to the technical scheme of the embodiment of the disclosure, the parallel corpus pair set is obtained, the two general translation models are trained respectively to obtain the trained general translation models respectively, the source language corpus set in the specific field forms the pseudo parallel corpus pair set by using one general translation model, and then the translation model in the target field is obtained according to the pseudo parallel corpus pair set and the other general translation model, so that the corresponding language translation model can be still established under the condition that the parallel corpus pair in the specific field is lacked, and the corpus translation accuracy in the specific field is greatly improved.
Optionally, on the basis of the foregoing technical solution, the first target translation model obtaining module 240 is further configured to:
and directionally training the trained first general translation model through the first pseudo parallel corpus pair set and the parallel corpus pair set to obtain a first target translation model.
Optionally, on the basis of the above technical solution, the apparatus for obtaining a translation model further includes:
the first convergence requirement judging module is used for judging whether the first target translation model meets the convergence requirement; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample; if the first target translation model does not meet the convergence requirement, the first target translation model is trained continuously through the first pseudo parallel corpus pair set until the first target translation model converges.
Optionally, on the basis of the above technical solution, the apparatus for obtaining a translation model further includes:
the second pseudo parallel corpus pair acquisition module is used for acquiring a second source language corpus set of a specific field, acquiring a second target language corpus set matched with the second source language corpus set of the specific field through the trained first universal translation model, and forming the second source language corpus set of the specific field and the matched second target language corpus set into a second pseudo parallel corpus pair set;
and the second target translation model acquisition module is used for directionally training the trained second general translation model through the second pseudo parallel corpus pair set to acquire a second target translation model.
Optionally, on the basis of the above technical solution, the apparatus for obtaining a translation model further includes:
the second convergence requirement judging module is used for judging whether the second target translation model meets the convergence requirement; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample; and if the second target translation model does not meet the convergence requirement, continuing training the second target translation model through the second pseudo parallel corpus pair set until the second target translation model converges.
Optionally, on the basis of the above technical solution, the initial training execution module 220 is specifically configured to:
performing initial training on a plurality of second general translation models through the parallel corpus pair set to obtain a plurality of trained second general translation models; wherein each of the second common translation models is of a different type.
Optionally, on the basis of the foregoing technical solution, the first pseudo-parallel corpus pair obtaining module 230 is specifically configured to:
and acquiring a first source language corpus set of a specific field, acquiring a plurality of matched first target language corpus sets through each trained second universal translation model, and forming a plurality of first pseudo-parallel corpus pair sets by the first source language corpus set of the specific field and each matched first target language corpus set.
Optionally, on the basis of the foregoing technical solution, the first target translation model obtaining module 240 is specifically configured to:
and directionally training the trained first general translation model through each first pseudo parallel corpus pair set to obtain a first target translation model.
Optionally, on the basis of the above technical solution, the apparatus for obtaining a translation model further includes:
a noise adding operation executing module, configured to perform a noise adding operation on the first pseudo-parallel corpus pair set to generate a first pseudo-parallel corpus pair set containing noise; the noise addition operation includes adding words, deleting words, or disordering the order of words in the corpus.
Optionally, on the basis of the above technical solution, the apparatus for obtaining a translation model further includes:
and the word order reversal operation execution module is used for respectively carrying out word order reversal operation on the first source language corpus set and the first target language corpus set in the first pseudo-parallel corpus pair set so as to obtain a new first pseudo-parallel corpus pair set.
The device can execute the method for acquiring the translation model provided by any embodiment of the disclosure, and has the corresponding functional modules and beneficial effects of the execution method. Technical details that are not elaborated in this embodiment may be referred to a method provided by any embodiment of the present disclosure.
EXAMPLE III
Fig. 3 shows a schematic structural diagram of an electronic device (e.g., the terminal device or the server in fig. 1) 300 suitable for implementing an embodiment of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, a mobile terminal such as a mobile phone, a notebook computer, a digital broadcast receiver, a PDA (personal digital assistant), a PAD (tablet computer), a PMP (portable multimedia player), a vehicle terminal (e.g., a car navigation terminal), and the like, and a stationary terminal such as a digital TV, a desktop computer, and the like. The electronic device shown in fig. 3 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present disclosure.
As shown in fig. 3, the electronic device 300 may include a processing means (e.g., a central processing unit, a graphics processor, etc.) 301 that may perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM)302 or a program loaded from a storage means 308 into a Random Access Memory (RAM) 303. In the RAM 303, various programs and data necessary for the operation of the electronic apparatus 300 are also stored. The processing device 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input/output (I/O) interface 305 is also connected to bus 304.
Generally, the following devices may be connected to the I/O interface 305: input devices 306 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; an output device 307 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage devices 308 including, for example, magnetic tape, hard disk, etc.; and a communication device 309. The communication means 309 may allow the electronic device 300 to communicate wirelessly or by wire with other devices to exchange data. While fig. 3 illustrates an electronic device 300 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a non-transitory computer readable medium, the computer program containing program code for performing the method illustrated by the flow chart. In such an embodiment, the computer program may be downloaded and installed from a network through the communication means 309, or installed from the storage means 308, or installed from the ROM 302. The computer program, when executed by the processing device 301, performs the above-described functions defined in the methods of the embodiments of the present disclosure.
It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may interconnect with any form or medium of digital data communication (e.g., a communications network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: acquiring a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora; respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set to obtain a trained first general translation model and a trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model; acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo parallel corpus pair set; and directionally training the trained first general translation model through the first pseudo parallel corpus pair set to obtain a first target translation model.
Computer program code for carrying out operations for the present disclosure may be written in any combination of one or more programming languages, including but not limited to an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present disclosure may be implemented by software or hardware. The name of a module does not constitute a limitation on the module itself in some cases, for example, the first target translation model obtaining module may be described as "a module for performing directional training on the trained first generic translation model through the first pseudo-parallel corpus pair set to obtain the first target translation model". The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), systems on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), and the like.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
According to one or more embodiments of the present disclosure, [ example 1 ] there is provided a translation model acquisition method including:
acquiring a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora;
respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set to obtain a trained first general translation model and a trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model;
acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo parallel corpus pair set;
and directionally training the trained first general translation model through the first pseudo parallel corpus pair set to obtain a first target translation model.
In accordance with one or more embodiments of the present disclosure, [ example 2 ] there is provided the method of example 1, further comprising:
and directionally training the trained first general translation model through the first pseudo parallel corpus pair set and the parallel corpus pair set to obtain a first target translation model.
In accordance with one or more embodiments of the present disclosure, [ example 3 ] there is provided the method of example 1, further comprising:
judging whether the first target translation model meets the convergence requirement or not; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample;
if the first target translation model does not meet the convergence requirement, the first target translation model is trained continuously through the first pseudo parallel corpus pair set until the first target translation model converges.
In accordance with one or more embodiments of the present disclosure, [ example 4 ] there is provided the method of example 1, further comprising:
acquiring a second source language corpus set of a specific field, acquiring a second target language corpus set matched with the second source language corpus set of the specific field through the trained first universal translation model, and forming a second pseudo parallel corpus pair set by the second source language corpus set of the specific field and the matched second target language corpus set;
and directionally training the trained second universal translation model through the second pseudo parallel corpus pair set to obtain a second target translation model.
In accordance with one or more embodiments of the present disclosure, [ example 5 ] there is provided the method of example 4, further comprising:
judging whether the second target translation model meets the convergence requirement or not; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample;
and if the second target translation model does not meet the convergence requirement, continuing training the second target translation model through the second pseudo parallel corpus pair set until the second target translation model converges.
In accordance with one or more embodiments of the present disclosure, [ example 6 ] there is provided the method of example 1, further comprising:
performing initial training on a plurality of second general translation models through the parallel corpus pair set to obtain a plurality of trained second general translation models; wherein each of the second generic translation models is of a different type;
acquiring a first source language corpus set of a specific field, acquiring a plurality of matched first target language corpus sets through each trained second universal translation model, and forming the first source language corpus set of the specific field and each matched first target language corpus set into a plurality of first pseudo-parallel corpus pair sets;
and directionally training the trained first general translation model through each first pseudo parallel corpus pair set to obtain a first target translation model.
In accordance with one or more embodiments of the present disclosure, [ example 7 ] there is provided the method of example 1, further comprising:
performing noise adding operation on the first pseudo parallel corpus pair set to generate a first pseudo parallel corpus pair set containing noise; the noise addition operation includes adding words, deleting words, and/or disordering the order of words in the corpus.
In accordance with one or more embodiments of the present disclosure, [ example 8 ] there is provided the method of example 1, further comprising:
and respectively performing word order reversal operation on a first source language corpus set and a first target language corpus set in the first pseudo-parallel corpus pair set to obtain a new first pseudo-parallel corpus pair set.
According to one or more embodiments of the present disclosure, [ example 9 ] there is provided an acquisition apparatus of a translation model, including:
the parallel corpus pair set acquisition module is used for acquiring a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora;
the initial training execution module is used for respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set so as to obtain the trained first general translation model and the trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model;
the first pseudo parallel corpus pair acquisition module is used for acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo parallel corpus pair set;
and the first target translation model acquisition module is used for directionally training the trained first general translation model through the first pseudo parallel corpus pair set to acquire a first target translation model.
According to one or more embodiments of the present disclosure, [ example 10 ] there is provided an electronic device comprising a memory, a processing apparatus, and a computer program stored on the memory and executable on the processing apparatus, the processing apparatus implementing the method of obtaining a translation model according to any one of examples 1-8 when executing the program.
According to one or more embodiments of the present disclosure, [ example 11 ] there is provided a storage medium containing computer-executable instructions for performing the method of obtaining a translation model according to any one of examples 1-8 when executed by a computer processor.
The foregoing description is only exemplary of the preferred embodiments of the disclosure and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the disclosure herein is not limited to the particular combination of features described above, but also encompasses other embodiments in which any combination of the features described above or their equivalents does not depart from the spirit of the disclosure. For example, the above features and (but not limited to) the features disclosed in this disclosure having similar functions are replaced with each other to form the technical solution.
Further, while operations are depicted in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order. Under certain circumstances, multitasking and parallel processing may be advantageous. Likewise, while several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the disclosure. Certain features that are described in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination.
Although the subject matter has been described in language specific to structural features and/or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the specific features and acts described above are disclosed as example forms of implementing the claims.
Claims (10)
1. A method for acquiring a translation model is characterized by comprising the following steps:
acquiring a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora;
respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set to obtain a trained first general translation model and a trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model; the trained second universal translation model is trained and completed based on vocabulary association and a grammar structure;
acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo parallel corpus pair set;
directionally training the trained first general translation model through the first pseudo parallel corpus pair set to obtain a first target translation model;
after the initial training is respectively performed on the first general translation model and the second general translation model through the parallel corpus pair set to obtain the trained first general translation model and second general translation model, the method further comprises the following steps:
acquiring a second source language corpus set of a specific field, acquiring a second target language corpus set matched with the second source language corpus set of the specific field through the trained first universal translation model, and forming a second pseudo parallel corpus pair set by the second source language corpus set of the specific field and the matched second target language corpus set;
and directionally training the trained second universal translation model through the second pseudo parallel corpus pair set to obtain a second target translation model.
2. The method according to claim 1, wherein directionally training the trained first generic translation model through the first set of pseudo-parallel corpus pairs to obtain a first target translation model comprises:
and directionally training the trained first general translation model through the first pseudo parallel corpus pair set and the parallel corpus pair set to obtain a first target translation model.
3. The method of claim 1, after obtaining the first target translation model, further comprising;
judging whether the first target translation model meets the convergence requirement or not; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample;
if the first target translation model does not meet the convergence requirement, the first target translation model is trained continuously through the first pseudo parallel corpus pair set until the first target translation model converges.
4. The method of claim 1, after obtaining the second target translation model, further comprising;
judging whether the second target translation model meets the convergence requirement or not; wherein the convergence requirement comprises that the translation result of the detection sample is matched with the parallel corpus of the detection sample;
and if the second target translation model does not meet the convergence requirement, continuing training the second target translation model through the second pseudo parallel corpus pair set until the second target translation model converges.
5. The method according to claim 1, wherein initially training the first and second generic translation models respectively through the set of parallel corpus pairs to obtain trained first and second generic translation models comprises:
performing initial training on a plurality of second general translation models through the parallel corpus pair set to obtain a plurality of trained second general translation models; wherein each of the second generic translation models is of a different type;
correspondingly, a first source language corpus set of a specific field is obtained, a first target language corpus set matched with the first source language corpus set of the specific field is obtained through the trained second universal translation model, and the first source language corpus set of the specific field and the matched first target language corpus set form a first pseudo parallel corpus pair set, which includes:
acquiring a first source language corpus set of a specific field, acquiring a plurality of matched first target language corpus sets through each trained second universal translation model, and forming the first source language corpus set of the specific field and each matched first target language corpus set into a plurality of first pseudo-parallel corpus pair sets;
correspondingly, the directional training of the trained first general translation model is performed through the first pseudo-parallel corpus pair set to obtain a first target translation model, including:
and directionally training the trained first general translation model through each first pseudo parallel corpus pair set to obtain a first target translation model.
6. The method according to claim 1, further comprising, after combining said first set of source language corpora of said particular domain and said first set of matching target language corpora into a first set of pseudo-parallel corpora pairs:
performing noise adding operation on the first pseudo parallel corpus pair set to generate a first pseudo parallel corpus pair set containing noise; the noise addition operation includes adding words, deleting words, and/or disordering the order of words in the corpus.
7. The method according to claim 1, further comprising, after combining said first set of source language corpora of said particular domain and said first set of matching target language corpora into a first set of pseudo-parallel corpora pairs:
and respectively performing word order reversal operation on a first source language corpus set and a first target language corpus set in the first pseudo-parallel corpus pair set to obtain a new first pseudo-parallel corpus pair set.
8. An apparatus for obtaining a translation model, comprising:
the parallel corpus pair set acquisition module is used for acquiring a parallel corpus pair set; wherein the set of parallel corpus pairs comprises at least one parallel corpus pair, each of the parallel corpus pairs comprising paired source and target language corpora;
the initial training execution module is used for respectively carrying out initial training on the first general translation model and the second general translation model through the parallel corpus pair set so as to obtain the trained first general translation model and the trained second general translation model; wherein the source language of the first generic translation model is the target language of the second generic translation model, and the target language of the first generic translation model is the source language of the second generic translation model; the trained second universal translation model is trained and completed based on vocabulary association and a grammar structure;
the first pseudo parallel corpus pair acquisition module is used for acquiring a first source language corpus set of a specific field, acquiring a first target language corpus set matched with the first source language corpus set of the specific field through the trained second universal translation model, and forming the first source language corpus set of the specific field and the matched first target language corpus set into a first pseudo parallel corpus pair set;
the first target translation model acquisition module is used for directionally training the trained first general translation model through the first pseudo parallel corpus pair set to acquire a first target translation model;
the second pseudo parallel corpus pair acquisition module is used for acquiring a second source language corpus set of a specific field, acquiring a second target language corpus set matched with the second source language corpus set of the specific field through the trained first universal translation model, and forming the second source language corpus set of the specific field and the matched second target language corpus set into a second pseudo parallel corpus pair set;
and the second target translation model acquisition module is used for directionally training the trained second general translation model through the second pseudo parallel corpus pair set to acquire a second target translation model.
9. An electronic device comprising a memory, a processing means and a computer program stored on the memory and executable on the processing means, characterized in that the processing means, when executing the program, implements the method of obtaining a translation model according to any of claims 1-7.
10. A storage medium containing computer-executable instructions for performing the method of obtaining a translation model according to any one of claims 1 to 7 when executed by a computer processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911251122.7A CN111008533B (en) | 2019-12-09 | 2019-12-09 | Method, device, equipment and storage medium for obtaining translation model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911251122.7A CN111008533B (en) | 2019-12-09 | 2019-12-09 | Method, device, equipment and storage medium for obtaining translation model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111008533A CN111008533A (en) | 2020-04-14 |
| CN111008533B true CN111008533B (en) | 2021-07-23 |
Family
ID=70115149
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911251122.7A Active CN111008533B (en) | 2019-12-09 | 2019-12-09 | Method, device, equipment and storage medium for obtaining translation model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111008533B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111859994B (en) * | 2020-06-08 | 2024-01-23 | 北京百度网讯科技有限公司 | Machine translation model acquisition and text translation method, device and storage medium |
| CN111859995B (en) * | 2020-06-16 | 2024-01-23 | 北京百度网讯科技有限公司 | Training method and device of machine translation model, electronic equipment and storage medium |
| CN111738025B (en) * | 2020-08-20 | 2020-11-17 | 腾讯科技（深圳）有限公司 | Artificial intelligence based translation method and device, electronic equipment and storage medium |
| CN112270200B (en) * | 2020-11-11 | 2023-10-27 | 北京有竹居网络技术有限公司 | Text information translation method and device, electronic equipment and storage medium |
| CN112380883B (en) * | 2020-12-04 | 2023-07-25 | 北京有竹居网络技术有限公司 | Model training method, machine translation method, device, equipment and storage medium |
| CN112784611A (en) * | 2021-01-21 | 2021-05-11 | 阿里巴巴集团控股有限公司 | Data processing method, device and computer storage medium |
| CN113204977B (en) * | 2021-04-29 | 2023-09-26 | 北京有竹居网络技术有限公司 | Information translation method, device, equipment and storage medium |
| CN113204979A (en) * | 2021-05-28 | 2021-08-03 | 北京金山数字娱乐科技有限公司 | Model training method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108563640A (en) * | 2018-04-24 | 2018-09-21 | 中译语通科技股份有限公司 | A kind of multilingual pair of neural network machine interpretation method and system |
| US10157347B1 (en) * | 2013-11-04 | 2018-12-18 | Predii, Inc. | Adaptable systems and methods for processing enterprise data |
| CN110334361A (en) * | 2019-07-12 | 2019-10-15 | 电子科技大学 | A kind of neural machine translation method towards rare foreign languages language |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106484682B (en) * | 2015-08-25 | 2019-06-25 | 阿里巴巴集团控股有限公司 | Machine translation method, device and electronic equipment based on statistics |
| CN109388809A (en) * | 2017-08-10 | 2019-02-26 | 陈虎 | It is a kind of for establishing the circulating transfer learning method of word translation model |
| CN108829684A (en) * | 2018-05-07 | 2018-11-16 | 内蒙古工业大学 | A kind of illiteracy Chinese nerve machine translation method based on transfer learning strategy |
| CN108920473B (en) * | 2018-07-04 | 2022-08-09 | 中译语通科技股份有限公司 | Data enhancement machine translation method based on same-class word and synonym replacement |
| CN110147450B (en) * | 2019-05-06 | 2021-08-03 | 北京科技大学 | Knowledge complementing method and device for knowledge graph |
-
2019
- 2019-12-09 CN CN201911251122.7A patent/CN111008533B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10157347B1 (en) * | 2013-11-04 | 2018-12-18 | Predii, Inc. | Adaptable systems and methods for processing enterprise data |
| CN108563640A (en) * | 2018-04-24 | 2018-09-21 | 中译语通科技股份有限公司 | A kind of multilingual pair of neural network machine interpretation method and system |
| CN110334361A (en) * | 2019-07-12 | 2019-10-15 | 电子科技大学 | A kind of neural machine translation method towards rare foreign languages language |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111008533A (en) | 2020-04-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111008533B (en) | Method, device, equipment and storage medium for obtaining translation model | |
| CN111046677B (en) | Method, device, equipment and storage medium for obtaining translation model | |
| CN111382261B (en) | Abstract generation method and device, electronic equipment and storage medium | |
| CN112270200B (en) | Text information translation method and device, electronic equipment and storage medium | |
| CN113139391B (en) | Translation model training method, device, equipment and storage medium | |
| CN112417902A (en) | Text translation method, device, equipment and storage medium | |
| CN111597825B (en) | Voice translation method and device, readable medium and electronic equipment | |
| CN113313064A (en) | Character recognition method and device, readable medium and electronic equipment | |
| WO2022228221A1 (en) | Information translation method, apparatus and device, and storage medium | |
| CN111368560A (en) | Text translation method and device, electronic equipment and storage medium | |
| CN113378586B (en) | Speech translation method, translation model training method, device, medium, and apparatus | |
| CN112712801A (en) | Voice wake-up method and device, electronic equipment and storage medium | |
| CN112306235A (en) | Gesture operation method, device, equipment and storage medium | |
| CN114170342A (en) | Image processing method, device, equipment and storage medium | |
| CN112380883B (en) | Model training method, machine translation method, device, equipment and storage medium | |
| CN111339789B (en) | Translation model training method and device, electronic equipment and storage medium | |
| CN112257459B (en) | Language translation model training method, translation method, device and electronic equipment | |
| WO2022121859A1 (en) | Spoken language information processing method and apparatus, and electronic device | |
| CN115129845A (en) | Text information processing method and device and electronic equipment | |
| CN114765025A (en) | Method for generating and recognizing speech recognition model, device, medium and equipment | |
| CN111221424B (en) | Method, apparatus, electronic device, and computer-readable medium for generating information | |
| CN114429629A (en) | Image processing method and device, readable storage medium and electronic equipment | |
| CN112489652A (en) | Text acquisition method and device for voice information and storage medium | |
| CN112418233A (en) | Image processing method, image processing device, readable medium and electronic equipment | |
| CN111581455A (en) | Text generation model generation method and device and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |