CN108595443A - Simultaneous interpreting method, device, intelligent vehicle mounted terminal and storage medium - Google Patents

Abstract

The present invention relates to the technical field of smart cars, and provides a simultaneous translation method and device, an intelligent vehicle terminal and a storage medium. The method includes: obtaining a voice to be translated based on a simultaneous translation request; performing voice recognition on the voice to be translated, and obtaining the voice to be translated Translate the text; identify the language of the text to be translated, and obtain the language to be translated corresponding to the text to be translated; according to the preset correspondence between the language to be translated and the target language, translate the text to be translated into the target language text, and voice to output. The invention realizes the automatic mutual translation between the two languages by automatically identifying the language of the speech to be translated, so that the people in the car who cannot speak the same language can communicate normally.

Description

Simultaneous translation method, device, intelligent vehicle-mounted terminal and storage medium

technical field

The invention relates to the technical field of smart cars, in particular to a simultaneous translation method, device, smart vehicle terminal and storage medium.

Background technique

Today's car functions are not just traditional transportation functions. With the development of the times and the continuous development of new technologies, cars are becoming more and more intelligent. As Chinese people continue to go abroad, foreigners continue to come to China for tourism and business. Often in taxis and commercial vehicles, drivers and personnel are unable to communicate normally because they do not understand the language of both parties when picking up passengers or receiving business.

Contents of the invention

The purpose of the embodiments of the present invention is to provide a simultaneous translation method, device, intelligent vehicle-mounted terminal and storage medium to solve the problem that drivers and passengers cannot communicate normally due to language barriers.

In order to achieve the above object, the technical solution adopted in the embodiment of the present invention is as follows:

In the first aspect, the embodiment of the present invention provides a simultaneous translation method, the method comprising: based on the simultaneous translation request, obtaining the speech to be translated; performing speech recognition on the speech to be translated to obtain the text to be translated; Identify and obtain the language to be translated corresponding to the text to be translated; translate the text to be translated into the target language text according to the preset correspondence between the language to be translated and the target language, and output it in the form of voice.

In the second aspect, the embodiment of the present invention also provides a simultaneous translation device, which includes an acquisition module, a speech recognition module, a language recognition module and a translation module. Among them, the acquisition module is used to obtain the speech to be translated based on the simultaneous translation request; the speech recognition module is used to perform speech recognition on the speech to be translated to obtain the text to be translated; the language recognition module is used to perform language recognition on the text to be translated to obtain the text to be translated The language to be translated corresponding to the text; the translation module is used to translate the text to be translated into a text in the target language according to the preset corresponding relationship between the language to be translated and the target language, and output it in the form of voice.

In the third aspect, the embodiment of the present invention also provides a kind of intelligent vehicle-mounted terminal, described intelligent vehicle-mounted terminal comprises vehicle-mounted microphone, vehicle-mounted sounder, and described intelligent vehicle-mounted terminal also comprises: memory; Processor, described processor and vehicle-mounted microphone , the vehicle-mounted sounders are all electrically connected; and a simultaneous translation device, the simultaneous translation device is stored in the memory and includes one or more software function modules executed by the processor, which includes: an acquisition module, It is used to obtain the voice to be translated based on the simultaneous translation request; the voice recognition module is used to perform voice recognition on the voice to be translated to obtain the text to be translated; the language recognition module is used to identify the voice to be translated Performing language identification on the translated text to obtain the language to be translated corresponding to the text to be translated; the translation module is used to translate the text to be translated into a text in the target language according to the preset correspondence between the language to be translated and the target language , and output in voice through the vehicle sounder.

In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above simultaneous translation method is implemented.

Compared with the prior art, the embodiment of the present invention provides a simultaneous translation method, device, intelligent vehicle-mounted terminal and storage medium. When the user needs to perform simultaneous translation, it responds to the simultaneous translation translation request issued by the user and obtains the speech to be translated. , and then conduct speech recognition on the speech to be translated to obtain the text to be translated, and then determine the language to be translated corresponding to the text to be translated by performing language recognition on the text to be translated, and then follow the preset correspondence between the language to be translated and the target language, Translate the text to be translated into the target language text and output it in voice mode. Compared with the prior art, the embodiment of the present invention realizes automatic simultaneous translation between two languages by automatically identifying the language of the speech to be translated, so that people in the car who cannot speak the same language can communicate normally.

In order to make the above objects, features and advantages of the present invention more comprehensible, the following specific embodiments are described in detail in conjunction with the accompanying drawings.

Description of drawings

In order to illustrate the technical solutions of the embodiments of the present invention more clearly, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and thus It should be regarded as a limitation on the scope, and those skilled in the art can also obtain other related drawings based on these drawings without creative work.

Fig. 1 shows a schematic block diagram of an intelligent vehicle terminal provided by an embodiment of the present invention.

Fig. 2 shows a flowchart of a simultaneous translation method provided by an embodiment of the present invention.

FIG. 3 is a flow chart of sub-steps of step S102 shown in FIG. 2 .

FIG. 4 is a flow chart of sub-steps of step S103 shown in FIG. 2 .

Fig. 5 shows a schematic block diagram of a simultaneous translation device provided by an embodiment of the present invention.

Icons: 100-intelligent vehicle terminal; 101-memory; 102-storage controller; 103-processor; 104-peripheral interface; 105-vehicle microphone; 106-vehicle sounder; 107-display device; 200-simultaneous translation device; 201-acquisition module; 202-speech recognition module; 203-language recognition module; 204-translation module; 205-display module.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. The components of the embodiments of the invention generally described and illustrated in the figures herein may be arranged and designed in a variety of different configurations. Accordingly, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts belong to the protection scope of the present invention.

It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish descriptions, and cannot be understood as indicating or implying relative importance.

Please refer to FIG. 1 , which shows a schematic block diagram of a smart vehicle terminal 100 provided by an embodiment of the present invention. The intelligent vehicle-mounted terminal 100 can be used to realize simultaneous translation between different languages, and can be a smart phone, a vehicle-mounted computer, a combination instrument or a multimedia host on a vehicle, and the like. The intelligent vehicle-mounted terminal 100 includes a memory 101 , a storage controller 102 , a processor 103 , a peripheral interface 104 , a vehicle-mounted microphone 105 , a vehicle-mounted sounder 106 , and a display device 107 .

The elements of the memory 101 , the memory controller 102 and the processor 103 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. The simultaneous translation device 200 includes at least one software function module that can be stored in the memory 101 in the form of software or firmware (firmware) or solidified in the operating system (operating system, OS) of the smart vehicle terminal 100 . The processor 103 is used to execute executable modules stored in the memory 101 , such as software function modules and computer programs included in the simultaneous translation device 200 .

Wherein, memory 101 can be, but not limited to, random access memory (Random Access Memory, RAM), read-only memory (Read Only Memory, ROM), programmable read-only memory (Programmable Read-OnlyMemory, PROM), erasable In addition to read-only memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable read-only memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc. Wherein, the memory 101 is used to store a program, and the processor 103 executes the program after receiving an execution instruction.

The processor 103 may be an integrated circuit chip with signal processing capability. Above-mentioned processor 103 can be general-purpose processor, comprises central processing unit (Central Processing Unit, CPU), network processor (NetworkProcessor, NP), speech processor and video processor etc.; Integrated circuits, field programmable gate arrays or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components. Various methods, steps and logic block diagrams disclosed in the embodiments of the present invention may be implemented or executed. The general-purpose processor may be a microprocessor, or the processor 103 may be any conventional processor or the like.

The vehicle-mounted microphone 105 is used to collect the voice of the simultaneous translation request and the voice to be translated, and send the voice to be translated to the intelligent vehicle-mounted terminal 100, so that the intelligent vehicle-mounted terminal 100 turns on the simultaneous translation function and realizes the translation between two different languages. Simultaneous translation. The on-vehicle microphone 105 is a microphone for a car, and is an energy conversion device that converts a sound signal into an electric signal. The vehicle-mounted microphone 105 may be a vehicle-mounted microphone, a vehicle-mounted microphone, a vehicle-mounted microphone, etc. In the embodiment of the present invention, the vehicle-mounted microphone 105 may be a vehicle-mounted microphone.

The vehicle-mounted sound generator 106 is a generator used in a car, and is a device that converts electrical energy into sound, and is used to output a response voice to a simultaneous translation request and a translated voice. The vehicle-mounted sound generator 106 may be a vehicle-mounted horn, a vehicle-mounted speaker, etc. In the embodiment of the present invention, the vehicle-mounted sound generator 106 may be a vehicle-mounted speaker.

The display device 107 is a kind of man-machine interface device on the car, which is used to display the user interface of the intelligent vehicle terminal 100, and can simultaneously receive the user's simultaneous translation request through touch and display the text to be translated and the translated text corresponding to the voice to be translated. The text in the target language is displayed, and the display device 107 may be a touch screen.

first embodiment

Please refer to FIG. 2 , which shows a flowchart of a simultaneous translation method provided by an embodiment of the present invention. The simultaneous translation method of the first embodiment of the present invention is applied to the intelligent vehicle terminal 100, and the simultaneous translation method includes the following steps:

Step S101, based on the simultaneous translation request, the speech to be translated is acquired.

In the embodiment of the present invention, the simultaneous translation request is a command request triggered by the user and used to start the simultaneous translation function realized by the simultaneous translation method, wherein the simultaneous translation request can be sent by the user through the vehicle-mounted microphone 105 The voice command may also be an operation command sent by the user through the user interface of the smart vehicle terminal 100 , or a keying command issued by the user through the buttons on the steering wheel of the car electrically connected to the smart vehicle terminal 100 . After the intelligent vehicle-mounted terminal 100 responds to the simultaneous translation request, it enters the simultaneous translation function and sends prompt information to the user. The prompt information can be a voice prompt sent by the vehicle-mounted sounder 106, or a display prompt sent by the display device 107, for example, the vehicle-mounted voice The device 106 broadcasts "please input voice", or the display device 107 displays "please input voice".

After the intelligent vehicle-mounted terminal 100 enters the simultaneous translation function, it starts to obtain the voice to be translated sent by the vehicle-mounted microphone 105, wherein the voice to be translated is the in-vehicle voice collected by the vehicle-mounted microphone 105, and the in-vehicle voice can be the sound emitted by the vehicle personnel when speaking. It may also be the sound produced by the sound-generating device in the car, for example, the sound played by an audio file on a mobile phone.

Step S102, performing speech recognition on the speech to be translated to obtain the text to be translated.

In the embodiment of the present invention, the purpose of speech recognition is to convert the speech uttered by the user into a text of character sequences readable by the smart vehicle terminal 100 . The method for performing speech recognition on the speech to be translated may include:

First, the speech to be translated is preprocessed to eliminate the influence of noise, and the acoustic features of the speech to be translated are extracted to obtain the audio data to be translated. Acoustic feature extraction can not only compress the information of the speech to be translated, but also facilitate subsequent speech recognition.

Secondly, input the audio data to be translated into the pre-established audio recognition model for processing to obtain the text to be translated, first calculate the probability that the speech to be translated corresponds to a syllable, obtain the syllable sequence of the speech to be translated, and then calculate the The probability of the corresponding word sequence, and finally select the word sequence with the highest syllable sequence probability and word sequence probability in the speech to be translated as the result of speech recognition, that is, the text to be translated.

Please refer to FIG. 3, step S102 may also include the following sub-steps:

Sub-step S1021, converting the speech to be translated into audio data to be translated.

In the embodiment of the present invention, firstly, speech enhancement is performed on the speech to be translated by eliminating noise and channel distortion. Secondly, the speech to be translated after speech enhancement is divided into frames, and Fourier transform is performed on each frame to extract the feature vector of each frame. The feature vector of each frame of the speech to be translated constitutes the audio data to be translated. For example, the speech to be translated is "Hello", and two speech frames of "You" and "Good" are obtained after frame division, then the audio data to be translated may include the feature vector of "You" and the feature vector of "Good".

Sub-step S1022, input the audio data to be translated into the pre-established audio recognition model, so as to obtain the text to be translated corresponding to the audio data to be translated.

In the embodiment of the present invention, the audio recognition model includes an acoustic model, a language model, and a search space, wherein the acoustic model is obtained after statistical modeling of the acoustic features of a large number of sample speeches, and is used to calculate the corresponding syllables of the speech to be translated. The probability. The language model is obtained by training a large number of text samples and using the method of probability and statistics to model the internal statistical laws of words, and is used to calculate the probability of the corresponding word sequence based on multiple syllable sequences.

The search space is a syllable-level network composed of syllables as nodes. The establishment process of the search space is as follows: First, the words that may appear in the text to be translated are used as nodes to form a word-level network. Among them, the words that may appear in the text to be translated can be It is pre-determined according to the application scenario of the speech to be translated. For example, if the application scenario is business reception, common words include hotels, countries, cities, etc.; The level-level network is the search space. For example, if the word network is "hotel", the expanded syllable-level network can be the syllables corresponding to "ho" and the syllables corresponding to "tel".

The process of inputting the audio data to be translated into the pre-established audio recognition model for speech recognition is: input the audio data to be translated into the search space, determine the word sequence with the highest probability in the search space according to the acoustic model and the phonetic model, and use the word The sequence serves as the text to be translated corresponding to the audio data to be translated.

Step S103, performing language identification on the text to be translated to obtain the language to be translated corresponding to the text to be translated.

In the embodiment of the present invention, after the text to be translated is obtained, firstly, the language features of the text to be translated are extracted to obtain the language features of the text to be translated. Among them, language features refer to unique signs that can be distinguished from other languages, usually including special letters, unique letter combinations, types of diacritics and the number of signs, etc. Feature extraction is to extract the unique letters, unique letter combinations, types of diacritics and the number of signs that are repeated and appear in large quantities in the text to be translated. The intelligent vehicle-mounted terminal 100 has a language database stored in advance, and multiple language templates and template languages corresponding to each language template are stored in the language database. After obtaining the language features of the text to be translated, the language features of the text to be translated and the language features Multiple language templates in the database are matched one by one to obtain the language to be translated corresponding to the text to be translated.

Please refer to FIG. 4, step S103 may also include the following sub-steps:

Sub-step S1031, extracting features of the text to be translated to obtain language features of the text to be translated.

In the embodiment of the present invention, language features refer to unique signs that can be distinguished from other languages, and generally include unique letters, unique letter combinations, types of diacritics, number of signs, and the like. For example, if the text to be translated is "Een van dehoofdkenmerken van dit communicationsysteem is dat het alleen door de mens kanworden voortgebracht en gebruikt en meestal ook alleen door de mens wordtbegrepen", then the extracted language features are "oo, ee, en". Sub-step S1032, matching the language feature with multiple language templates in the language database one by one to determine the language to be translated corresponding to the language feature.

In the embodiment of the present invention, a language database is pre-stored on the intelligent vehicle-mounted terminal 100, and the language database includes multiple language templates and template languages corresponding to each language template. For example, the language templates are "oo, aa, uu, ee", the template language corresponding to this language template is Dutch. The language features of the text to be translated obtained in substep S1031 are matched one by one with multiple language templates in the language database, and the template language corresponding to the target language template whose matching degree reaches a predetermined threshold is used as the language to be translated corresponding to the text to be translated, for example, the language to be translated The translated language feature is "oo, ee, uu", the language template is "oo, aa, uu, ee", the matching degree is (3/4)*100%=75%, and 4 refers to the features in the language template The number is 4, and 3 means that the number of language features to be translated consistent with the features in the language template is 3, and the predetermined threshold is an empirical value. For example, based on accuracy considerations, the predetermined threshold is 70%. Find the template language corresponding to the target language template with the highest matching degree from the matching degree greater than or equal to the predetermined threshold as the language to be translated corresponding to the text to be translated, for example, the predetermined threshold is 70%, and the text to be translated is equal to the Dutch language template The matching degree is 72%, and the matching degree between the text to be translated and the language template of Afrikaans is 80%, so Afrikaans is used as the language to be translated corresponding to the text to be translated.

As an implementation, the smart vehicle-mounted terminal 100 can also communicate with a language database server, on which a language database is stored, and the smart vehicle-mounted terminal 100 sends the language characteristics of the text to be translated to the language database server, so that The language database server performs language feature matching, and sends the matching result to the smart vehicle terminal 100 .

Step S104: Translate the text to be translated into text in the target language according to the preset correspondence between the language to be translated and the target language, and output it in a voice manner.

In the embodiment of the present invention, the user pre-sets the corresponding relationship between the language to be translated and the target language through the operation interface of the smart vehicle terminal 100, and the language to be translated corresponding to the text to be translated obtained in step S103 and the preset language to be translated The corresponding relationship with the target language can obtain the target language of the target language text. For example, the corresponding relationship set by the user is Chinese-English mutual translation. When the language to be translated is Chinese, the target language of the target language text is English. At this time, the text to be translated needs to be translated into English text. When the language to be translated is English, The target language of the target language text is Chinese. At this time, the text to be translated needs to be translated into Chinese text.

In the embodiment of the present invention, the process of translating the text to be translated into the text in the target language can be realized by inputting the text to be translated into the pre-established cyclic neural network model to obtain the text in the target language corresponding to the text to be translated. The cyclic neural network model includes two stages of encoding and decoding. First, the text to be translated is input into the cyclic neural network model for encoding to obtain the vector sequence corresponding to the text to be translated. Then, the vector sequence is decoded, and by calculating the target language text The probability of each word in the target language text is finally obtained.

As an implementation, the process of inputting the text to be translated into the pre-established cyclic neural network model to obtain the target language text corresponding to the text to be translated may be:

First, the text to be translated is encoded to obtain a vector sequence corresponding to the text to be translated. Input each word in the text to be translated into the recurrent neural network model in turn for vector conversion to obtain the word vector corresponding to each word, wherein the word vector corresponding to each word is used as the input to generate the word vector of the next word. In this way, The word vector of the last word actually contains the word vector information of all words before the word, so the word vector of the last word is used as the vector sequence corresponding to the text to be translated.

Secondly, the vector sequence corresponding to the text to be translated is decoded, and the probability of each word in the target language text is calculated one by one, and the target language text is finally obtained. Decoding is to decode the vector sequence corresponding to the text to be translated. Each time a word in the target language text is decoded, and when decoding each word, the previous word that has been decoded must be used as input, and then each word is determined to be in the target language. The probability in the language text, each time the word with the highest probability is selected as the word in the target language text, and finally the target language text is obtained.

It should be noted that the cyclic neural network model in step S104 is pre-established through training, and the process of training the cyclic neural network model is as follows:

First, initialize the input parameters in the recurrent neural network model. The input parameters include embedding size, encoder input direction, encoding depth, decoding depth, neural unit type, and number of iterations. Among them, the embedding size is used to represent the word vector length, that is, the word vector length; the encoder input direction refers to the input to be The order of the input sequence of the translated text can be forward, reverse, or forward and reverse. The forward direction is input in sequence according to the order of the text to be translated. The order and reverse order of the text to be translated are input in sequence; the depth of encoding refers to the number of layers of the encoding neural network, and the depth of decoding refers to the number of layers of the decoding neural network; commonly used neural unit types include simple recurrent neural network, long short-term memory network, gating Recurrent unit, after determining the type of neural unit, at the same time initialize the set size of the neural network corresponding to the neural unit, the number of neurons in each layer, the activation function of each layer and the initial value of the gradient weight; the number of iterations mainly affects the training loss of the neural network value.

Second, input the training samples into the recurrent neural network model for training and continuously update the weight parameters and offsets. Each training sample includes a sample text to be translated and a corresponding target language sample text, for example, {(where are you going?), (Where are you going?)} is a training sample, and the sample text to be translated is (you Where are you going?), the corresponding target language sample text is (Where are you going?). Each recurrent neural network includes at least one input layer, one hidden layer, and one output layer, and the weight parameters include: weight parameters from the input layer to the hidden layer; weight parameters from the hidden layer to the hidden layer; The weight parameter of the output layer, the offset includes the offset of the hidden layer and the offset of the output layer. Input the training samples into the cyclic neural network model, use the backpropagation algorithm BPTT to train the cyclic neural network model, and use the gradient descent algorithm to update the weight parameters and offsets, and finally obtain the values of the weight parameters and offsets, In this way, a recurrent neural network model that can translate the text to be translated into the target language text is trained.

In the embodiment of the present invention, after the smart vehicle-mounted terminal 100 obtains the text in the target language, it can output the text in the target language in a voice mode. Sounder 106 output. In addition, the user may not hear the audio in the target language in time due to noise, etc. In order to improve user experience, the text to be translated and the translated text in the target language can be displayed on the display device 107. Therefore, the embodiment of the present invention may also include Step S105.

Step S105, displaying both the text to be translated and the text in the target language corresponding to the text to be translated.

In the embodiment of the present invention, in addition to converting the text in the target language into audio in the target language and outputting it through the vehicle-mounted sounder 106, the smart vehicle-mounted terminal 100 can also convert the text to be translated and the text corresponding to the text to be translated through the vehicle-mounted display device 107. All texts in the target language are displayed, so that when the user does not hear the audio in the target language due to noise, the user can see the text in the target language through the display device 107, thereby improving user experience.

Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

First of all, the voice to be translated is obtained based on the user's simultaneous translation request. The simultaneous translation request can be voice, operation, or button in various forms, which is convenient for users to initiate requests in different scenarios.

Secondly, use speech recognition to convert the speech to be translated into the text to be translated, and then perform automatic language recognition on the text to be translated, realizing automatic simultaneous translation between the two languages, so that people in the car who do not understand the language can communicate normally .

Third, output the text in the target language after the speech translation to be translated in a voice mode, or display the text in the target language corresponding to the text to be translated and the text to be translated with the display device 107, through two forms of voice output and text display The output meets the output needs of users in different scenarios and improves the user experience.

second embodiment

Please refer to FIG. 5 , which shows a schematic block diagram of a simultaneous translation device 200 provided by an embodiment of the present invention. The simultaneous translation device 200 is applied to the intelligent vehicle-mounted terminal 100 , which includes an acquisition module 201 ; a speech recognition module 202 ; a language recognition module 203 ; a translation module 204 ; and a display module 205 .

The obtaining module 201 is configured to obtain the speech to be translated based on the simultaneous translation request.

The voice recognition module 202 is configured to perform voice recognition on the voice to be translated to obtain the text to be translated.

In the embodiment of the present invention, the speech recognition module 202 is specifically used to convert the speech to be translated into audio data to be translated; input the audio data to be translated into a pre-established audio recognition model to obtain the text to be translated corresponding to the audio data to be translated.

The language identification module 203 is configured to identify the language of the text to be translated to obtain the language to be translated corresponding to the text to be translated.

In the embodiment of the present invention, the language identification module 203 is specifically used to perform feature extraction on the text to be translated to obtain the language features of the text to be translated; to match the language features with multiple language templates in the language database one by one to determine the corresponding language features The languages to be translated.

The translation module 204 is configured to translate the text to be translated into text in the target language according to the preset correspondence between the language to be translated and the target language, and output the text in a voice manner.

The display module 205 is configured to display both the text to be translated and the text in the target language corresponding to the text to be translated.

The embodiment of the present invention also discloses a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by the processor 103, the simultaneous translation method disclosed in the foregoing embodiments of the present invention is realized.

In summary, the present invention provides a simultaneous translation method, device, intelligent vehicle-mounted terminal, and storage medium, the method comprising: obtaining the speech to be translated based on the simultaneous translation request; performing speech recognition on the speech to be translated, and obtaining the speech to be translated Translate the text; identify the language of the text to be translated, and obtain the language to be translated corresponding to the text to be translated; according to the preset correspondence between the language to be translated and the target language, translate the text to be translated into the target language text, and voice to output. Compared with the prior art, the embodiment of the present invention realizes automatic simultaneous translation between two languages by automatically identifying the language of the speech to be translated, so that people in the car who cannot speak the same language can communicate normally.

In the several embodiments provided in this application, it should be understood that the disclosed devices and methods may also be implemented in other ways. The device embodiments described above are only illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functions and possible implementations of devices, methods and computer program products according to multiple embodiments of the present invention. operate. In this regard, each block in a flowchart or block diagram may represent a module, program segment, or part of code that includes one or more Executable instructions. 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 in succession may, in fact, be executed substantially concurrently, or they may sometimes be executed in the reverse order, depending upon the functionality involved. It should also be noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented by a dedicated hardware-based system that performs the specified function or action , or may be implemented by a combination of dedicated hardware and computer instructions.

In addition, each functional module in each embodiment of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

If the functions are realized in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including Several instructions are used to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes. . It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is a relationship between these entities or operations. There is no such actual relationship or order between them. Furthermore, the term "comprises", "comprises" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, but also includes elements not expressly listed. other elements of or also include elements inherent in such a process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article or apparatus comprising said element.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and changes. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention. It should be noted that like numerals and letters denote similar items in the following figures, therefore, once an item is defined in one figure, it does not require further definition and explanation in subsequent figures.

Claims (10)

1. A simultaneous translation method, characterized in that it is applied to an intelligent vehicle-mounted terminal, and the method comprises: Obtain the voice to be translated based on the simultaneous translation request; performing speech recognition on the speech to be translated to obtain the text to be translated; performing language identification on the text to be translated to obtain the language to be translated corresponding to the text to be translated; According to the preset correspondence between the language to be translated and the target language, the text to be translated is translated into a text in the target language, and output in a voice manner. 2. The method according to claim 1, wherein the step of performing speech recognition on the speech to be translated to obtain the text to be translated comprises: converting the speech to be translated into audio data to be translated; Inputting the audio data to be translated into a pre-established audio recognition model to obtain a text to be translated corresponding to the audio data to be translated. 3. The method according to claim 1, wherein the intelligent vehicle-mounted terminal includes a language database, a plurality of language templates are stored in the language database, and the language identification is carried out to the text to be translated to obtain the The steps of describing the language to be translated corresponding to the text to be translated include: performing feature extraction on the text to be translated to obtain language features of the text to be translated; The language feature is matched one by one with multiple language templates in the language database to determine the language to be translated corresponding to the language feature. 4. The method according to claim 1, wherein the step of translating the text to be translated into a target language text includes: Inputting the text to be translated into a pre-established cyclic neural network model to obtain a text in the target language corresponding to the text to be translated. 5. The method of claim 1, further comprising: Both the text to be translated and the text in the target language corresponding to the text to be translated are displayed. 6. A simultaneous translation device, characterized in that it is applied to an intelligent vehicle-mounted terminal, and the device comprises: An acquisition module, configured to acquire the voice to be translated based on the simultaneous translation request; A voice recognition module, configured to perform voice recognition on the voice to be translated to obtain the text to be translated; A language identification module, configured to identify the language of the text to be translated, and obtain the language to be translated corresponding to the text to be translated; The translation module is configured to translate the text to be translated into text in the target language according to the preset correspondence between the language to be translated and the target language, and output the text in a voice manner. 7. The device according to claim 6, wherein the speech recognition module is specifically used for: converting the speech to be translated into audio data to be translated; Inputting the audio data to be translated into a pre-established audio recognition model to obtain a text to be translated corresponding to the audio data to be translated. 8. The device according to claim 6, wherein the intelligent vehicle-mounted terminal includes a language database, and a plurality of language templates are stored in the language database. The language identification module is specifically used for: performing feature extraction on the text to be translated to obtain language features of the text to be translated; The language feature is matched one by one with multiple language templates in the language database to determine the language to be translated corresponding to the language feature. 9. A kind of intelligent vehicle-mounted terminal, it is characterized in that, described intelligent vehicle-mounted terminal comprises vehicle-mounted microphone, vehicle-mounted sounder, and described intelligent vehicle-mounted terminal also comprises: memory; Processor, described processor is electrically connected with vehicle-mounted microphone, vehicle-mounted sounder; And A simultaneous translation device, the simultaneous translation device is stored in the memory and includes one or more software function modules executed by the processor, including: An acquisition module, configured to acquire the speech to be translated collected by the vehicle-mounted microphone based on the simultaneous translation request; A voice recognition module, configured to perform voice recognition on the voice to be translated to obtain the text to be translated; A language identification module, configured to identify the language of the text to be translated, and obtain the language to be translated corresponding to the text to be translated; The translation module is used for translating the text to be translated into text in the target language according to the preset correspondence between the language to be translated and the target language, and outputting the text in a voice manner through the vehicle-mounted sounder. 10. A computer-readable storage medium, on which a computer program is stored, wherein, when the computer program is executed by a processor, the method according to any one of claims 1-5 is implemented.