CN112133279A - Vehicle-mounted information broadcasting method and device and terminal equipment - Google Patents
Vehicle-mounted information broadcasting method and device and terminal equipment Download PDFInfo
- Publication number
- CN112133279A CN112133279A CN201910492587.5A CN201910492587A CN112133279A CN 112133279 A CN112133279 A CN 112133279A CN 201910492587 A CN201910492587 A CN 201910492587A CN 112133279 A CN112133279 A CN 112133279A
- Authority
- CN
- China
- Prior art keywords
- information
- text
- abstract
- voice
- target
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012545 processing Methods 0.000 claims abstract description 93
- 238000013528 artificial neural network Methods 0.000 claims abstract description 75
- 230000002457 bidirectional effect Effects 0.000 claims abstract description 24
- 230000007246 mechanism Effects 0.000 claims abstract description 24
- 238000003062 neural network model Methods 0.000 claims description 26
- 238000001228 spectrum Methods 0.000 claims description 26
- 230000015572 biosynthetic process Effects 0.000 claims description 21
- 238000003786 synthesis reaction Methods 0.000 claims description 21
- 238000004590 computer program Methods 0.000 claims description 19
- 239000013598 vector Substances 0.000 claims description 12
- 125000004122 cyclic group Chemical group 0.000 claims description 9
- 238000011176 pooling Methods 0.000 claims description 8
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000002194 synthesizing effect Effects 0.000 claims description 7
- 238000000605 extraction Methods 0.000 description 10
- 230000006870 function Effects 0.000 description 9
- 230000008569 process Effects 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 238000004140 cleaning Methods 0.000 description 4
- 238000004891 communication Methods 0.000 description 4
- 230000004044 response Effects 0.000 description 4
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000009432 framing Methods 0.000 description 3
- 230000000306 recurrent effect Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 230000004913 activation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/34—Browsing; Visualisation therefor
- G06F16/345—Summarisation for human users
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L13/00—Speech synthesis; Text to speech systems
- G10L13/02—Methods for producing synthetic speech; Speech synthesisers
- G10L13/04—Details of speech synthesis systems, e.g. synthesiser structure or memory management
- G10L13/047—Architecture of speech synthesisers
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L13/00—Speech synthesis; Text to speech systems
- G10L13/08—Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Databases & Information Systems (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention is suitable for the technical field of vehicle-mounted systems, and provides a method, a device and terminal equipment for broadcasting vehicle-mounted information, wherein the method comprises the following steps: acquiring target information, wherein the target information is information received by target terminal equipment; obtaining text abstract information corresponding to the target information according to the target information and a text processing neural network, wherein the text processing neural network is a bidirectional circulation neural network based on an attention mechanism; obtaining corresponding information abstract voice according to the text abstract information; and broadcasting the information abstract voice, or sending the information abstract voice to broadcasting equipment. The embodiment of the invention can improve the driving safety while transmitting information in time.
Description
Technical Field
The invention belongs to the technical field of vehicle-mounted systems, and particularly relates to a vehicle-mounted information broadcasting method, a vehicle-mounted information broadcasting device and terminal equipment.
Background
With the development of economy and technology, automobiles are more and more popular in people's daily life. In the current information era, a driver needs to timely acquire information received by terminal equipment such as a mobile phone and the like even when driving, so that the problem of how to timely acquire the information of the terminal equipment by the driver on the premise of ensuring driving safety exists when driving an automobile.
In the prior art, a device capable of completely converting text information received by a terminal device into voice to be played to a driver is generally added in an automobile to solve the problem. However, in the prior art, the voice playing content is often too long, so that the attention of the driver is greatly dispersed, and the driving safety threat exists.
Disclosure of Invention
In view of this, embodiments of the present invention provide a method, an apparatus, and a terminal device for broadcasting vehicle-mounted information, so as to solve the problem in the prior art how to improve driving safety while timely conveying information.
A first aspect of an embodiment of the present invention provides a method for broadcasting vehicle-mounted information, including:
acquiring target information, wherein the target information is information received by target terminal equipment;
obtaining text abstract information corresponding to the target information according to the target information and a text processing neural network, wherein the text processing neural network is a bidirectional circulation neural network based on an attention mechanism;
obtaining corresponding information abstract voice according to the text abstract information;
and broadcasting the information abstract voice, or sending the information abstract voice to broadcasting equipment.
A second aspect of the embodiments of the present invention provides a vehicle-mounted information broadcasting device, including:
the target information acquisition unit is used for acquiring target information, wherein the target information is information received by target terminal equipment;
the text abstract information acquisition unit is used for acquiring text abstract information corresponding to the target information according to the target information and a text processing neural network, wherein the text processing neural network is a bidirectional cyclic neural network based on an attention mechanism;
the voice synthesis unit is used for obtaining corresponding information abstract voice according to the text abstract information;
and the broadcasting unit is used for broadcasting the information abstract voice or sending the information abstract voice to broadcasting equipment.
A third aspect of the embodiments of the present invention provides a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the vehicle-mounted information broadcasting method when executing the computer program.
A fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, where a computer program is stored, and when the computer program is executed by a processor, the steps of the vehicle-mounted information broadcasting method are implemented.
Compared with the prior art, the embodiment of the invention has the following beneficial effects: in the embodiment of the invention, after the target information of the target terminal equipment is received, the target information is abstracted and extracted through the bidirectional circulation neural network model based on the attention mechanism, the text abstract information corresponding to the target information can be accurately obtained, and the text abstract information is converted into the information abstract voice to be played.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed for the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.
Fig. 1 is a schematic flow chart illustrating an implementation of a first vehicle-mounted information broadcasting method according to an embodiment of the present invention;
FIG. 2 is a flow chart of data processing of a neural network model for speech processing according to an embodiment of the present invention;
FIG. 3 is a schematic flow chart of extracting a summary of text information through a text processing neural network according to an embodiment of the present invention;
FIG. 4 is a flow chart of synthesizing information-summarized voice through a voice synthesis neural network according to an embodiment of the present invention;
fig. 5 is a schematic flow chart illustrating an implementation of a second method for broadcasting vehicle-mounted information according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a vehicle-mounted information broadcasting device according to an embodiment of the present invention;
fig. 7 is a schematic diagram of a terminal device according to an embodiment of the present invention.
Detailed Description
In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.
In order to explain the technical means of the present invention, the following description will be given by way of specific examples.
It will be understood that the terms "comprises" and/or "comprising," when used in this specification and the appended claims, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof.
It is also to be understood that the terminology used in the description of the present application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification of the present application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
As used in this specification and the appended claims, the term "if" may be interpreted contextually as "when", "upon" or "in response to a determination" or "in response to a detection". Similarly, the phrase "if it is determined" or "if a [ described condition or event ] is detected" may be interpreted contextually to mean "upon determining" or "in response to determining" or "upon detecting [ described condition or event ]" or "in response to detecting [ described condition or event ]".
In addition, in the description of the present application, the terms "first," "second," "third," and the like are used solely to distinguish one from another and are not to be construed as indicating or implying relative importance.
The first embodiment is as follows:
fig. 1 shows a schematic flow diagram of a first vehicle-mounted information broadcasting method provided in an embodiment of the present application, which is detailed as follows:
in S101, target information is obtained, where the target information is information received by a target terminal device.
The target information is information received by the target terminal equipment, and comprises short messages, message pushing, voice information, news pushing and the like. The target terminal device is an intelligent terminal device capable of receiving information through a wireless network, for example, a mobile phone terminal carried by a driver, and the target terminal device may also be a terminal device such as a notebook computer and a tablet computer, which is not limited herein. And if the target terminal equipment is detected to generate information interaction, acquiring the target information received by the target terminal equipment. Optionally, before the obtaining the target information received by the target terminal device, the method further includes: and establishing a communication connection channel with the target terminal equipment in a wired or wireless mode.
In S102, text summary information corresponding to the target information is obtained according to the target information and a text processing neural network, where the text processing neural network is a bidirectional cyclic neural network based on an attention mechanism.
And performing abstract extraction processing on the target information through a text processing neural network so as to obtain text abstract information corresponding to the target information. The text processing Neural Network is a Bidirectional Recurrent Neural Network (BRNN) based on Attention Mechanism (Attention Mechanism). The Bidirectional circulation neural network is specifically a Bidirectional circulation Long-Short Term Memory network (BilSTM), and the target information is abstracted through the BilSTM, so that information correlation in two directions of the context of the target information can be analyzed simultaneously, and the accuracy of abstract extraction is improved. Meanwhile, an attention mechanism is added in the BilSTM network, so that different weights are given to each data in the target information sequence when the data are processed, the data with higher weights can be weighted during abstract extraction, and the accuracy of abstract extraction is further improved.
Specifically, after the target information is acquired, a data type of the target information is detected, where the data type may be a text data type or a voice data type.
Specifically, if the target information is text information, the text information is input into a text processing neural network, and text abstract information corresponding to the target information is obtained.
The text processing neural network based on the attention mechanism bidirectional cyclic neural network is a sequence-to-sequence encoding and decoding neural network model, and obtains an abstract text corresponding to an input text by encoding and decoding the input text. Therefore, if the target information is text information of a text data type, the target information is directly input into the text processing neural network, a series of feature extraction and operation processes are carried out, and text abstract information corresponding to the target information is output and obtained.
Specifically, if the target information is voice information, inputting the voice information into a voice processing module to obtain text information corresponding to the voice information;
and inputting the text information into a text processing neural network to obtain text abstract information corresponding to the target information.
If the target information is detected to be voice information of a voice data type, the voice information needs to be converted into text information of a corresponding text data type, and the text information can be input into a text processing neural network to extract text abstract information. And inputting the voice information into a voice processing module, and converting the voice information into corresponding text information according to the characteristic information of the voice information. And inputting the text information of the text data type corresponding to the target information of the original voice data type, which is obtained by conversion, into a text processing neural network to obtain text abstract information corresponding to the target information.
Optionally, the inputting the voice information into the voice processing module to obtain the text information corresponding to the voice information includes:
preprocessing the voice information to generate a spectrogram;
and inputting the spectrogram into a speech processing neural network model to obtain text information corresponding to the speech information, wherein the speech processing neural network model comprises a plurality of convolution layers, a pooling layer and a softmax classifier layer.
The voice information preprocessing comprises the processing of data cleaning, framing processing, Fourier transformation and the like on voice. Carrying out data cleaning on the voice information, and eliminating useless information such as noise, blank sections in audio and the like in the voice information to obtain voice data which only contains effective voice basically; performing framing, windowing and other processing on the whole section of voice data after data cleaning, and dividing the voice data into short audio frequencies of one frame and one frame, wherein the duration of each short audio frequency frame can be 10 ms-30 ms; and performing Fourier transform on the short audio subjected to frame division processing frame by frame to respectively obtain a frequency spectrum corresponding to each frame of short audio, and splicing and stacking the frequency spectrums corresponding to each frame of short audio according to the time sequence of the short audio to obtain a spectrogram with the characteristic information of the whole section of voice data. The voice data is subjected to data cleaning to remove invalid data, so that the influence of interference signals can be reduced, and the subsequent voice processing efficiency can be improved; meanwhile, as the voice belongs to a quasi-steady signal, namely, the voice has short-time stationarity, when the voice signal is processed, in order to reduce the unsteady and time-varying influence of a longer whole section of voice data, the voice data is subjected to framing processing and then Fourier transform, so that the spectral feature information of the voice data can be more effectively extracted.
And inputting the spectrogram of the frequency spectrum characteristic information containing the voice information obtained after the preprocessing into a voice processing neural network model to obtain text information corresponding to the voice information. The speech processing neural network model is a trained end-to-end neural network model containing a multilayer convolution structure and comprises a plurality of convolution layers, a plurality of pooling layers and a softmax classifier. The data processing process of the speech processing neural network model is shown in fig. 2, and is detailed as follows:
s1: and inputting the voice spectrum X corresponding to the spectrogram into a voice processing neural network model, and preliminarily extracting voice spectrum characteristics through two convolution layers to obtain first voice characteristic information.
S2: and carrying out average pooling on the first voice characteristic information through a first pooling layer to obtain optimized second voice characteristic information.
S3: and performing further feature extraction on the second voice feature information through another convolution layer to obtain third voice feature information.
S4: performing dimensionality reduction processing on the third voice feature information through a second pooling layer to obtain a final feature vector;
s5: and performing probability calculation on the feature vector through a softmax classifier, outputting a text sequence, and obtaining text information finally corresponding to the spectrogram.
Because the voice information is usually longer, namely the corresponding spectrogram is usually longer, the voice spectrogram corresponding to the voice information is processed by the voice processing neural network model comprising the multilayer convolution structure, so that the network can be ensured to learn more information content, and the features in the long spectrogram are extracted as comprehensively as possible; meanwhile, the voice processing neural network model can be more stable in the training process through multiple convolution and pooling operations, so that the performance of the voice processing neural network model is better.
Optionally, the inputting the text information into a text processing neural network to obtain text summary information corresponding to the target information includes:
a1: inputting the text information into a text processing neural network to generate an original text sequence, and enabling the original text sequence to pass through a first attention layer to generate a first attention weight, wherein the text processing neural network is a bidirectional circulation neural network based on an attention mechanism and comprising the first attention layer and a second attention layer;
a2: obtaining a current abstract sequence according to all generated abstract words at present, and converting the current abstract sequence into corresponding word vectors;
a3: passing the first attention weight and the word vector through a second attention layer to obtain a second attention weight;
a4: decoding to obtain a next abstract word according to the original text sequence, the current abstract sequence and the second attention weight;
a5: and repeating the steps A2-A4 until the abstract words corresponding to the original text information are generated, and obtaining the text abstract information corresponding to the target information.
The text processing neural network is a bidirectional cyclic neural network based on an attention mechanism, and particularly relates to a sequence-to-sequence encoding and decoding neural network model. The processing procedure of the text information by the text processing neural network is specifically shown in fig. 3, and is detailed as follows:
at A1, the text information is input into the encoder of the neural network for text processing, the original text sequence is obtained by encoding the text information, and the original text sequence is input into the encoderAn attention layer passes through the first hidden layer and the first weight output layer to obtain a first attention weight W1. The first attention weight W1Which reflects the attention of each word to each other between the original text sequences.
In a2, a current digest sequence is obtained by assembling all generated digest words decoded by a decoder layer, and a word vector corresponding to the current digest sequence is obtained by performing operations on the current digest sequence through a word embedding layer, a second hiding layer, an activation function layer (preferably a tangent function tanh), and the like. If the first round of processing is currently performed, the word vector is a 0 matrix because no abstract word is currently generated, i.e., the current abstract sequence is blank.
In A3, a weighting operation is performed according to the obtained first attention weight and the word vector corresponding to the current abstract sequence, and the weighting operation is input into a third hidden layer, and a second attention weight W is obtained through a second weight output layer2. Second attention weight W2Which reflects the attention of the current abstract sequence to each word in the original text sequence.
At a4, the original text sequence, the current abstract sequence and the second attention weight obtained at A3 are input into the decoder, and a new abstract word is determined to be generated according to the classification probability P.
In a5, the steps a2 to a4 are repeatedly executed, each abstract word is generated one by one, if it is detected that the generation of the abstract word corresponding to the current original text information is finished, the step cycle is ended, and all the generated abstract words are assembled to obtain a complete abstract sentence, namely the text abstract information corresponding to the target information.
In the embodiment of the present invention, the first weight W is calculated by the first attention tier in step A11The attention of each word of the original text sequence is realized, namely the key words of the original text sequence are balanced; the weight W of the original text sequence generated according to step A11Generating a second weight W by a second attention layer weighting operation on the word vector corresponding to the current abstract sequence2The weight W2Indicating that the current abstract word is generated on the premise of the current abstract wordThe attention size of each word in the original text sequence is to be sequenced. Therefore, through the two attention layers, the text processing neural network can accurately extract the key words when the text abstract is extracted, so that the generation of the text abstract information is more accurate.
In S103, a corresponding message digest voice is obtained according to the text digest information.
And performing voice synthesis on the text abstract information of the text data type extracted in the step S102 to obtain corresponding information abstract voice, so that the original target information is converted into information in a voice format only containing brief key content.
Optionally, the step S103 specifically includes:
and inputting the text abstract information into a speech synthesis neural network to obtain corresponding information abstract speech, wherein the speech synthesis neural network comprises a convolutional network coding layer, a third attention layer, a convolutional network decoding layer, a bridging layer and an acoustic code layer.
The speech synthesis neural network is specifically an end-to-end neural network model trained in advance based on coding-decoding, and specifically includes a convolutional network coding layer, a third attention layer, a convolutional network decoding layer, a bridge layer and an acoustic code layer Vocoder, and a data processing flow thereof is shown in fig. 4.
Optionally, the inputting the text summary information into a speech synthesis neural network to obtain corresponding information summary speech includes:
the text abstract information sequentially passes through the convolutional network coding layer, the third attention layer and the convolutional network decoding layer to obtain a Mel logarithm spectrum corresponding to the text abstract information;
enabling the Mel logarithmic spectrum to pass through the bridging layer to obtain a linear logarithmic sound spectrum;
and synthesizing the abstract information voice by passing the linear logarithmic voice spectrum through the voice code layer.
S10301: and sequentially passing the text abstract information through the convolutional network coding layer, the third attention layer and the convolutional network decoding layer to obtain a Mel logarithm spectrum corresponding to the text abstract information.
Firstly, text abstract information is subjected to text coding and feature capture through a convolutional network coding layer, and hidden layer state output of the convolutional network coding layer is obtained; the hidden layer state output is subjected to weighted operation through a third attention layer to obtain the attention weight W for the attention size of each word in the text abstract information3(ii) a Attention is weighted by W3And the hidden layer state of the convolutional network encoder layer is output and input into the convolutional network decoder layer to obtain the Mel logarithm spectrum log-mel corresponding to the text abstract information.
S10302: and passing the Mel logarithmic spectrum through the bridging layer to obtain a linear logarithmic sound spectrum.
And inputting the log-mel sound spectrum obtained by decoding the convolutional network decoder layer into a bridging layer, wherein the bridging layer is specifically a convolution processing module, and learning is carried out through the bridging layer, the time sequence of an input sequence is sorted, and a linear log-linear sound spectrum is obtained by prediction, so that the voice synthesis effect is improved.
S10303: and synthesizing the abstract information voice by passing the linear logarithmic voice spectrum through the voice code layer.
And (4) enabling the log-linear sound spectrum to pass through a Vocoder to generate a sound spectrum waveform, and synthesizing an audio waveform signal corresponding to the abstract information voice.
And in S104, broadcasting the information summary voice or sending the information summary voice to broadcasting equipment.
And broadcasting the synthesized information abstract voice so that a driver can timely obtain the key contents in the target information without any manual operation. Specifically, the information summary voice can be broadcasted through the broadcasting unit of the vehicle-mounted information broadcasting device. Or, the broadcast device is instructed to broadcast the information summary voice by sending the information summary voice to other broadcast devices. Optionally, the broadcasting device may be a target terminal device, that is, information summary voice obtained by processing the target information acquired from the target terminal device in the above steps is returned to the target terminal device, so as to instruct the target terminal device to broadcast the information summary voice.
In the embodiment of the invention, after the target information of the target terminal equipment is received, the target information is abstracted and extracted through the bidirectional circulation neural network model based on the attention mechanism, the text abstract information corresponding to the target information can be accurately obtained, and the text abstract information is converted into the information abstract voice to be played.
Example two:
fig. 5 shows a schematic flow chart of a second vehicle-mounted information broadcasting method provided in the embodiment of the present application, which is detailed as follows:
in S501, a target application of a target terminal device is set.
The target terminal device is an intelligent terminal device capable of receiving information through a wireless network, for example, a mobile phone terminal carried by a driver, and the target terminal device may also be a terminal device such as a notebook computer, a tablet computer, and the like. The target terminal device supports various applications including an application capable of receiving information, such as a short message application for receiving short messages, a news application for receiving news push, a chat application for receiving voice messages, and the like. And receiving a setting instruction, and setting the target application of the target terminal equipment according to the setting instruction. The setting instruction is an instruction for a driver to specify a target application, and through setting the target application, the driver can specify the target application containing important information to be obtained in time and obtain the information of the target application in time.
In S502, if it is detected that the target application of the target terminal device receives the target information, the target information is acquired.
And if the target application of the target terminal equipment receives the target information, acquiring the target information, wherein the target information is information such as short messages, message push, voice information or news push received by the specified target application. And acquiring the target information, triggering and executing the data processing process from the step S503 to the step S505, and broadcasting the key contents in the target information to the driver in time. And if the current target terminal equipment receives the interactive information but the information is not the target information received by the target application, ignoring and not acquiring the information.
In S503, obtaining text summary information corresponding to the target information according to the target information and a text processing neural network, where the text processing neural network is a bidirectional cyclic neural network based on an attention mechanism.
In this embodiment, S503 is the same as S102 in the previous embodiment, and please refer to the related description of S102 in the previous embodiment, which is not repeated herein.
In S504, a corresponding message digest voice is obtained according to the text digest information.
In this embodiment, S504 is the same as S103 in the previous embodiment, and please refer to the related description of S103 in the previous embodiment, which is not repeated herein.
In S505, the message digest voice is broadcasted, or the message digest voice is sent to a broadcasting device.
In this embodiment, S505 is the same as S104 in the previous embodiment, and please refer to the related description of S104 in the previous embodiment, which is not repeated herein.
In the embodiment of the invention, the target application is set in advance, only the target information received by the target application is acquired, namely only the target information concerned by the driver is acquired, and the information except the target information is ignored, so that only the information concerned by the driver can be efficiently transmitted, the dispersion of other unimportant information to the attention of the driver is reduced, and the driving safety is further improved.
It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.
Example three:
fig. 6 is a schematic structural diagram of a vehicle-mounted information broadcasting device according to an embodiment of the present application, and for convenience of description, only parts related to the embodiment of the present application are shown:
this on-vehicle information broadcasts device includes: a target information acquisition unit 61, a text summary information acquisition unit 62, a voice synthesis unit 63, and a broadcast unit 64. Wherein:
the target information acquiring unit 61 is configured to acquire target information, where the target information is information received by a target terminal device.
The target information is information received by the target terminal equipment, and comprises short messages, message pushing, voice information, news pushing and the like. The target terminal device is an intelligent terminal device capable of receiving information through a wireless network, for example, a mobile phone terminal carried by a driver, and the target terminal device may also be a terminal device such as a notebook computer and a tablet computer, which is not limited herein. And if the target terminal equipment is detected to generate information interaction, acquiring the target information received by the target terminal equipment. Optionally, before the obtaining the target information received by the target terminal device, the method further includes: and establishing a communication connection channel with the target terminal equipment in a wired or wireless mode.
And a text summary information obtaining unit 62, configured to obtain text summary information corresponding to the target information according to the target information and a text processing neural network, where the text processing neural network is a bidirectional cyclic neural network based on an attention mechanism.
And performing abstract extraction processing on the target information through a text processing neural network so as to obtain text abstract information corresponding to the target information. The text processing Neural Network is a Bidirectional Recurrent Neural Network (BRNN) based on Attention Mechanism (Attention Mechanism). The Bidirectional circulation neural network is specifically a Bidirectional circulation Long-Short Term Memory network (BilSTM), and the target information is abstracted through the BilSTM, so that information correlation in two directions of the context of the target information can be analyzed simultaneously, and the accuracy of abstract extraction is improved. Meanwhile, an attention mechanism is added in the BilSTM network, so that each data in the sequence has different weight when the data is processed, the data with higher weight can be weighted when abstract extraction is carried out, and the accuracy of the abstract extraction is further improved.
Optionally, the text summary information obtaining unit 62 includes:
and the first text summary information acquisition module is used for inputting the text information into a text processing neural network if the target information is the text information to obtain the text summary information corresponding to the target information.
Optionally, the text summary information obtaining unit 62 includes a speech processing unit and a second text summary information obtaining module:
the voice processing unit is used for inputting the voice information into a voice processing module if the target information is the voice information to obtain text information corresponding to the voice information;
and the second text abstract information acquisition module is used for inputting the text information into a text processing neural network to obtain the text abstract information corresponding to the target information.
Optionally, the speech processing unit includes a preprocessing module and a speech processing neural network model:
the preprocessing module is used for preprocessing the voice information to generate a spectrogram;
and the voice processing neural network model is used for inputting the spectrogram into the voice processing neural network model to obtain text information corresponding to the voice information, wherein the voice processing neural network model comprises a plurality of convolution layers, a pooling layer and a softmax classifier layer.
Optionally, the text summary information obtaining unit 62 includes a bidirectional recurrent neural network model based on an attention mechanism, and is configured to: a1: inputting the text information into a text processing neural network to generate an original text sequence, and enabling the original text sequence to pass through a first attention layer to generate a first attention weight, wherein the text processing neural network is a bidirectional cyclic neural network model based on an attention mechanism and comprising the first attention layer and a second attention layer; a2: obtaining a current abstract sequence according to all generated abstract words at present, and converting the current abstract sequence into corresponding word vectors; a3: passing the first attention weight and the word vector through a second attention layer to obtain a second attention weight; a4: decoding to obtain a next abstract word according to the original text sequence, the current abstract sequence and the second attention weight; a5: and repeating the steps A2-A4 until the abstract words corresponding to the original text information are generated, and obtaining the text abstract information corresponding to the target information.
And the voice synthesis unit 63 is configured to obtain corresponding information summary voice according to the text summary information.
And carrying out voice synthesis on the extracted text abstract information of the text data type to obtain corresponding information abstract voice, so that the original target information is converted into the information of the voice format only containing brief key contents.
Optionally, the speech synthesis unit 63 comprises:
and the voice synthesis neural network module is used for inputting the text summary information into a voice synthesis neural network to obtain corresponding information summary voice, wherein the voice synthesis neural network comprises a convolutional network coding layer, a third attention layer, a convolutional network decoding layer, a bridging layer and an acoustic code layer.
Optionally, the speech synthesis neural network module is specifically configured to: the text abstract information sequentially passes through the convolutional network coding layer, the third attention layer and the convolutional network decoding layer to obtain a Mel logarithm spectrum corresponding to the text abstract information; enabling the Mel logarithmic spectrum to pass through the bridging layer to obtain a linear logarithmic sound spectrum; and synthesizing the abstract information voice by passing the linear logarithmic voice spectrum through the voice code layer.
And the broadcasting unit 64 is used for broadcasting the information abstract voice or sending the information abstract voice to broadcasting equipment.
And broadcasting the synthesized information abstract voice so that a driver can timely obtain the key contents in the target information without any manual operation. Specifically, the information abstract voice is directly broadcasted; or, the broadcast device is instructed to broadcast the information summary voice by sending the information summary voice to other broadcast devices. Optionally, the broadcasting device may be a target terminal device, that is, information summary voice obtained by processing target information acquired from the target terminal device is returned to the target terminal device, so as to instruct the target terminal device to broadcast the information summary voice.
Optionally, the on-vehicle information broadcasting device further includes:
a setting unit for setting a target application of a target terminal device;
correspondingly, the target information obtaining unit 61 is configured to obtain the target information if it is detected that the target application of the target terminal device receives the target information.
In the embodiment of the invention, after the target information of the target terminal equipment is received, the target information is abstracted and extracted through the bidirectional circulation neural network model based on the attention mechanism, the text abstract information corresponding to the target information can be accurately obtained, and the text abstract information is converted into the information abstract voice to be played.
Example four:
fig. 7 is a schematic diagram of a terminal device according to an embodiment of the present invention. As shown in fig. 7, the terminal device 7 of this embodiment includes: a processor 70, a memory 71, and a computer program 72, such as a telematics program, stored in the memory 71 and operable on the processor 70. The processor 70 implements the steps in each of the above-described embodiments of the in-vehicle information broadcasting method, such as steps S101 to S104 shown in fig. 1, when executing the computer program 72. Alternatively, the processor 70, when executing the computer program 72, implements the functions of the modules/units in the above-described device embodiments, such as the functions of the modules 61 to 64 shown in fig. 6.
Illustratively, the computer program 72 may be partitioned into one or more modules/units that are stored in the memory 71 and executed by the processor 70 to implement the present invention. The one or more modules/units may be a series of computer program instruction segments capable of performing specific functions, which are used to describe the execution process of the computer program 72 in the terminal device 7. For example, the computer program 72 may be divided into a target information obtaining unit, a text summary information obtaining unit, a speech synthesizing unit, and a broadcasting unit, and the specific functions of each unit are as follows:
and the target information acquisition unit is used for acquiring target information, wherein the target information is information received by the target terminal equipment.
And the text abstract information acquisition unit is used for acquiring the text abstract information corresponding to the target information according to the target information and a text processing neural network, wherein the text processing neural network is a bidirectional cyclic neural network model based on an attention mechanism.
And the voice synthesis unit is used for obtaining corresponding information abstract voice according to the text abstract information.
And the broadcasting unit is used for broadcasting the information abstract voice or sending the information abstract voice to broadcasting equipment.
The terminal device 7 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The terminal device may include, but is not limited to, a processor 70, a memory 71. It will be appreciated by those skilled in the art that fig. 7 is merely an example of a terminal device 7 and does not constitute a limitation of the terminal device 7 and may comprise more or less components than shown, or some components may be combined, or different components, for example the terminal device may further comprise input output devices, network access devices, buses, etc.
The Processor 70 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
The memory 71 may be an internal storage unit of the terminal device 7, such as a hard disk or a memory of the terminal device 7. The memory 71 may also be an external storage device of the terminal device 7, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the terminal device 7. Further, the memory 71 may also include both an internal storage unit and an external storage device of the terminal device 7. The memory 71 is used for storing the computer program and other programs and data required by the terminal device. The memory 71 may also be used to temporarily store data that has been output or is to be output.
It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated in one processing unit, or each unit may exist alone physically, or two or more units are integrated in one unit, and the integrated unit may be implemented in a form of hardware, or in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
In the above embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to the related descriptions of other embodiments for parts that are not described or illustrated in a certain embodiment.
Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other ways. For example, the above-described embodiments of the apparatus/terminal device are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions when actually implemented, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.
In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.
The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, all or part of the flow of the method according to the embodiments of the present invention may also be implemented by a computer program, which may be stored in a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the method embodiments may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain content that is subject to appropriate increase or decrease as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media does not include electrical carrier signals and telecommunications signals as is required by legislation and patent practice.
The above-mentioned embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.
Claims (10)
1. A vehicle-mounted information broadcasting method is characterized by comprising the following steps:
acquiring target information, wherein the target information is information received by target terminal equipment;
obtaining text abstract information corresponding to the target information according to the target information and a text processing neural network, wherein the text processing neural network is a bidirectional circulation neural network based on an attention mechanism;
obtaining corresponding information abstract voice according to the text abstract information;
and broadcasting the information abstract voice, or sending the information abstract voice to broadcasting equipment.
2. The method for broadcasting vehicle-mounted information according to claim 1, wherein the obtaining of the text summary information corresponding to the target information according to the target information and a text processing neural network comprises:
and if the target information is text information, inputting the text information into a text processing neural network to obtain text abstract information corresponding to the target information.
3. The method for broadcasting vehicle-mounted information according to claim 1, wherein the obtaining of the text summary information corresponding to the target information according to the target information and a text processing neural network comprises:
if the target information is voice information, inputting the voice information into a voice processing module to obtain text information corresponding to the voice information;
and inputting the text information into a text processing neural network to obtain text abstract information corresponding to the target information.
4. The method for broadcasting vehicle-mounted information according to claim 3, wherein the step of inputting the voice information into a voice processing module to obtain text information corresponding to the voice information comprises:
preprocessing the voice information to generate a spectrogram;
and inputting the spectrogram into a speech processing neural network model to obtain text information corresponding to the speech information, wherein the speech processing neural network model comprises a plurality of convolution layers, a pooling layer and a softmax classifier layer.
5. A method for broadcasting vehicle-mounted information according to claim 2 or 3, wherein the step of inputting the text information into a text processing neural network to obtain text summary information corresponding to the target information comprises:
a1: inputting the text information into a text processing neural network to generate an original text sequence, and enabling the original text sequence to pass through a first attention layer to generate a first attention weight, wherein the text processing neural network is a bidirectional circulation neural network based on an attention mechanism and comprising the first attention layer and a second attention layer;
a2: obtaining a current abstract sequence according to all generated abstract words at present, and converting the current abstract sequence into corresponding word vectors;
a3: passing the first attention weight and the word vector through a second attention layer to obtain a second attention weight;
a4: decoding to obtain a next abstract word according to the original text sequence, the current abstract sequence and the second attention weight;
a5: and repeating the steps A2-A4 until the abstract words corresponding to the original text information are generated, and obtaining the text abstract information corresponding to the target information.
6. The method for broadcasting vehicle-mounted information according to claim 1, wherein the obtaining of the corresponding information summary voice according to the text summary information includes:
and inputting the text abstract information into a speech synthesis neural network to obtain corresponding information abstract speech, wherein the speech synthesis neural network comprises a convolutional network coding layer, a third attention layer, a convolutional network decoding layer, a bridging layer and an acoustic code layer.
7. The method for broadcasting vehicle-mounted information according to claim 6, wherein the step of inputting the text summary information into a speech synthesis neural network to obtain corresponding information summary speech includes:
the text abstract information sequentially passes through the convolutional network coding layer, the third attention layer and the convolutional network decoding layer to obtain a Mel logarithm spectrum corresponding to the text abstract information;
enabling the Mel logarithmic spectrum to pass through the bridging layer to obtain a linear logarithmic sound spectrum;
and synthesizing the abstract information voice by passing the linear logarithmic voice spectrum through the voice code layer.
8. A vehicle-mounted information broadcasting method according to any one of claims 1 to 7, further comprising, before the acquiring the target information:
setting a target application of a target terminal device;
correspondingly, the acquiring target information includes:
and if the target application of the target terminal equipment is detected to receive the target information, acquiring the target information.
9. The utility model provides a vehicle carried information reports device which characterized in that includes:
the target information acquisition unit is used for acquiring target information, wherein the target information is information received by target terminal equipment;
the text abstract information acquisition unit is used for acquiring text abstract information corresponding to the target information according to the target information and a text processing neural network, wherein the text processing neural network is a bidirectional cyclic neural network based on an attention mechanism;
the voice synthesis unit is used for obtaining corresponding information abstract voice according to the text abstract information;
and the broadcasting unit is used for broadcasting the information abstract voice or sending the information abstract voice to broadcasting equipment.
10. A terminal device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of claims 1 to 8 when executing the computer program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910492587.5A CN112133279B (en) | 2019-06-06 | 2019-06-06 | Vehicle-mounted information broadcasting method and device and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910492587.5A CN112133279B (en) | 2019-06-06 | 2019-06-06 | Vehicle-mounted information broadcasting method and device and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112133279A true CN112133279A (en) | 2020-12-25 |
| CN112133279B CN112133279B (en) | 2024-06-21 |
Family
ID=73849096
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910492587.5A Active CN112133279B (en) | 2019-06-06 | 2019-06-06 | Vehicle-mounted information broadcasting method and device and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112133279B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116414972A (en) * | 2023-03-08 | 2023-07-11 | 浙江方正印务有限公司 | Method for automatically broadcasting information content and generating short message |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1554179A (en) * | 2001-09-13 | 2004-12-08 | 皇家飞利浦电子股份有限公司 | Summary extraction and preview of important information from voice messages |
| JP2010066675A (en) * | 2008-09-12 | 2010-03-25 | Fuji Xerox Co Ltd | Voice information processing system and voice information processing program |
| CN105405439A (en) * | 2015-11-04 | 2016-03-16 | 科大讯飞股份有限公司 | Voice playing method and device |
| CN106652995A (en) * | 2016-12-31 | 2017-05-10 | 深圳市优必选科技有限公司 | Text voice broadcasting method and system |
| CN107943405A (en) * | 2016-10-13 | 2018-04-20 | 广州市动景计算机科技有限公司 | Sound broadcasting device, method, browser and user terminal |
| CN108281139A (en) * | 2016-12-30 | 2018-07-13 | 深圳光启合众科技有限公司 | Speech transcription method and apparatus, robot |
| CN108763191A (en) * | 2018-04-16 | 2018-11-06 | 华南师范大学 | A kind of text snippet generation method and system |
| CN109472024A (en) * | 2018-10-25 | 2019-03-15 | 安徽工业大学 | A kind of file classification method based on bidirectional circulating attention neural network |
| CN109657051A (en) * | 2018-11-30 | 2019-04-19 | 平安科技（深圳）有限公司 | Text snippet generation method, device, computer equipment and storage medium |
| CN109697974A (en) * | 2017-10-19 | 2019-04-30 | 百度（美国）有限责任公司 | Use the system and method for the neural text-to-speech that convolution sequence learns |
| CN109726281A (en) * | 2018-12-12 | 2019-05-07 | Tcl集团股份有限公司 | A kind of text snippet generation method, intelligent terminal and storage medium |
| CN109767752A (en) * | 2019-02-27 | 2019-05-17 | 平安科技（深圳）有限公司 | A kind of phoneme synthesizing method and device based on attention mechanism |
-
2019
- 2019-06-06 CN CN201910492587.5A patent/CN112133279B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1554179A (en) * | 2001-09-13 | 2004-12-08 | 皇家飞利浦电子股份有限公司 | Summary extraction and preview of important information from voice messages |
| JP2010066675A (en) * | 2008-09-12 | 2010-03-25 | Fuji Xerox Co Ltd | Voice information processing system and voice information processing program |
| CN105405439A (en) * | 2015-11-04 | 2016-03-16 | 科大讯飞股份有限公司 | Voice playing method and device |
| CN107943405A (en) * | 2016-10-13 | 2018-04-20 | 广州市动景计算机科技有限公司 | Sound broadcasting device, method, browser and user terminal |
| CN108281139A (en) * | 2016-12-30 | 2018-07-13 | 深圳光启合众科技有限公司 | Speech transcription method and apparatus, robot |
| CN106652995A (en) * | 2016-12-31 | 2017-05-10 | 深圳市优必选科技有限公司 | Text voice broadcasting method and system |
| CN109697974A (en) * | 2017-10-19 | 2019-04-30 | 百度（美国）有限责任公司 | Use the system and method for the neural text-to-speech that convolution sequence learns |
| CN108763191A (en) * | 2018-04-16 | 2018-11-06 | 华南师范大学 | A kind of text snippet generation method and system |
| CN109472024A (en) * | 2018-10-25 | 2019-03-15 | 安徽工业大学 | A kind of file classification method based on bidirectional circulating attention neural network |
| CN109657051A (en) * | 2018-11-30 | 2019-04-19 | 平安科技（深圳）有限公司 | Text snippet generation method, device, computer equipment and storage medium |
| CN109726281A (en) * | 2018-12-12 | 2019-05-07 | Tcl集团股份有限公司 | A kind of text snippet generation method, intelligent terminal and storage medium |
| CN109767752A (en) * | 2019-02-27 | 2019-05-17 | 平安科技（深圳）有限公司 | A kind of phoneme synthesizing method and device based on attention mechanism |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116414972A (en) * | 2023-03-08 | 2023-07-11 | 浙江方正印务有限公司 | Method for automatically broadcasting information content and generating short message |
| CN116414972B (en) * | 2023-03-08 | 2024-02-20 | 浙江方正印务有限公司 | Method for automatically broadcasting information content and generating short message |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112133279B (en) | 2024-06-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110600059B (en) | Acoustic event detection method and device, electronic equipment and storage medium | |
| CN112233698B (en) | Character emotion recognition method, device, terminal equipment and storage medium | |
| CN108595443A (en) | Simultaneous interpreting method, device, intelligent vehicle mounted terminal and storage medium | |
| CN111667835A (en) | Voice recognition method, living body detection method, model training method and device | |
| CN115602165B (en) | Digital employee intelligent system based on financial system | |
| CN115116458B (en) | Voice data conversion method, device, computer equipment and storage medium | |
| CN113555032A (en) | Multi-speaker scene recognition and network training method and device | |
| CN114338623A (en) | Audio processing method, device, equipment, medium and computer program product | |
| CN107731232A (en) | Voice translation method and device | |
| CN111369968B (en) | Speech synthesis method and device, readable medium and electronic equipment | |
| WO2022005615A1 (en) | Speech enhancement | |
| CN112133279B (en) | Vehicle-mounted information broadcasting method and device and terminal equipment | |
| CN102571882A (en) | Network-based voice reminding method and system | |
| CN111863009B (en) | Training method and system of context information prediction model | |
| CN113113048A (en) | Speech emotion recognition method and device, computer equipment and medium | |
| CN114694654A (en) | Audio processing method and device, terminal equipment and computer readable storage medium | |
| CN116645956A (en) | Speech synthesis method, speech synthesis system, electronic device, and storage medium | |
| CN111477248A (en) | Audio noise detection method and device | |
| CN110852042A (en) | Character type conversion method and device | |
| CN115762557A (en) | Training method and system of self-supervision training predictor for voice separation | |
| CN113436621B (en) | GPU (graphics processing Unit) -based voice recognition method and device, electronic equipment and storage medium | |
| CN117063229A (en) | Interactive voice signal processing method, related equipment and system | |
| CN114049875A (en) | TTS (text to speech) broadcasting method, device, equipment and storage medium | |
| CN113870896A (en) | Motion sound false judgment method and device based on time-frequency graph and convolutional neural network | |
| CN117316160B (en) | Silent speech recognition method, silent speech recognition apparatus, electronic device, and computer-readable medium |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: 516006 TCL science and technology building, No. 17, Huifeng Third Road, Zhongkai high tech Zone, Huizhou City, Guangdong Province Applicant after: TCL Technology Group Co.,Ltd. Address before: 516006 Guangdong province Huizhou Zhongkai hi tech Development Zone No. nineteen District Applicant before: TCL Corp. Country or region before: China |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |