CN112289310A - Voice processing method and electronic equipment - Google Patents
Voice processing method and electronic equipment Download PDFInfo
- Publication number
- CN112289310A CN112289310A CN202011183800.3A CN202011183800A CN112289310A CN 112289310 A CN112289310 A CN 112289310A CN 202011183800 A CN202011183800 A CN 202011183800A CN 112289310 A CN112289310 A CN 112289310A
- Authority
- CN
- China
- Prior art keywords
- forwarding
- processing circuit
- slave processing
- forwarding data
- cyclic
- 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.)
- Withdrawn
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 8
- 238000000034 method Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims description 105
- 125000004122 cyclic group Chemical group 0.000 claims description 56
- 238000013528 artificial neural network Methods 0.000 claims description 22
- 230000000306 recurrent effect Effects 0.000 claims description 20
- 238000013473 artificial intelligence Methods 0.000 claims description 17
- 230000004913 activation Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 3
- 239000004984 smart glass Substances 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 16
- 239000011159 matrix material Substances 0.000 description 7
- 238000010586 diagram Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000005265 energy consumption Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 229940050561 matrix product Drugs 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 238000012827 research and development Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- 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
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/16—Speech classification or search using artificial neural networks
-
- 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
- G10L15/00—Speech recognition
- G10L15/28—Constructional details of speech recognition systems
Landscapes
- Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- User Interface Of Digital Computer (AREA)
Abstract
The application provides a voice processing method, which comprises the following steps: the electronic equipment processes the voice data to obtain the meaning of the voice. The technical scheme provided by the application has the advantages of reducing the calculation power consumption and improving the user experience.
Description
Technical Field
The present application relates to the field of speech, and in particular, to a speech processing method and an electronic device.
Background
In the prior art, artificial intelligence has been applied to many fields, such as speech recognition and the like. Neural networks in artificial intelligence have the largest potential at present, and most researchers put the research and development into the field reversely.
For the neural network, the existing artificial intelligence has large calculation amount and high cost when carrying out voice recognition.
Disclosure of Invention
The invention aims to provide a voice processing method, and the technical scheme can reduce the calculation overhead, reduce the power consumption and improve the user experience.
In a first aspect, a method for processing speech is provided, where the method is performed by an electronic device, and the electronic device includes: an artificial intelligence chip, the artificial intelligence chip structure comprising: the electronic device includes: an artificial intelligence chip, the artificial intelligence chip structure comprising: the main processing circuit, 2 multiplexer, 6 groups follow treatment circuit, every group follow treatment circuit includes: a plurality of slave processing circuits and 1 forwarding circuit, wherein the multiplexer switch is 4P 6T; the 6 ports of the main processing circuit are respectively connected with each forwarding circuit of the 6 groups of slave processing circuits, and each forwarding circuit is respectively connected with the broadcast ports of a plurality of slave processing circuits of the same group of slave processing circuits; the other 8 ports of the main processing circuit are respectively connected with 4P ports of 2 4P6T switches, and 6T ports of the 2 4P6T switches are respectively connected with a first slave processing circuit and a second slave processing circuit which are adjacent in each group of slave processing circuits; the slave processing circuit is also connected with other adjacent slave processing circuits in the same group of slave processing circuits through two forwarding ports; the method comprises the following steps:
the electronic equipment acquires a first voice, and the electronic equipment extracts input data X of the first voice at the t-1 momentt-1Inputting the data Xt-1Determining the input data as the input data of the recurrent neural network at the t-1 th moment;
the electronic equipment inputs data Xt-1Inputting the weight value W into a main processing circuit, calling the weight value W of the recurrent neural network by the main processing circuit, and executing t-1 layer operation to obtain a hidden layer output result S at t-1 momentt-1And t-1 layer output result Ot-1；
The main processing circuit receives the input data X of the first voice at the time tt(ii) a Mixing XtAnd St-1Performing an addition operation to obtain (X)t+St-1) (ii) a Will (X)t+St-1) Determining the data to be cyclic conversion data, determining the weight W to be broadcast forwarding data, cutting the broadcast forwarding data into a plurality of broadcast forwarding data blocks, respectively broadcasting the broadcast forwarding data blocks to a forwarding circuit through delta ports, cutting the cyclic forwarding data into alpha groups of cyclic forwarding data blocks, and sending the alpha groups of cyclic forwarding data blocks to a first slave processing circuit and a second slave processing circuit through a 4P6T switch;
the forwarding circuit forwards the received broadcast forwarding data block to a plurality of slave processing circuits in the same group of slave processing circuits; when the P port of the 4P6T switch receives a group of cyclic forwarding data blocks, one T port connected with one group of T ports is sent to the first slave processing circuit, and when the P port of the 4P6T switch receives another group of cyclic forwarding data blocks, the other T port connected with the other group of T ports is sent to the second slave processing circuit;
when the first slave processing circuit receives a group of cyclic forwarding data blocks, intercepting a local cyclic forwarding data block from the group of cyclic forwarding data blocks, and forwarding the rest cyclic forwarding data blocks to other slave processing circuits anticlockwise; when the second slave processing circuit receives another group of cyclic forwarding data blocks, intercepting local cyclic forwarding data blocks from the group of cyclic forwarding data blocks, and forwarding the rest cyclic forwarding data blocks to other slave processing circuits clockwise;
the slave processing circuit receives the residual cyclic forwarding data block through one forwarding port, receives the broadcast forwarding data block through the broadcast port, intercepts the local cyclic forwarding data block from the residual cyclic forwarding data block, and sends other cyclic forwarding data blocks to other adjacent slave processing circuits through another forwarding port; performing inner product operation on the local circulating forwarding data block and the broadcasting forwarding data block to obtain an operation result, and sending the operation result to a forwarding circuit through a broadcasting port;
the forwarding circuit forwards the operation result to the main processing circuit; the main processing circuit obtains the hidden layer output S of the t-th layer according to the operation resulttAnd the tth output result Ot。
Basis of artificial intelligence chip St、OtAnd executing subsequent operation of the t-th layer to obtain a result of the recurrent neural network, and obtaining the meaning of the first voice according to the result of the recurrent neural network.
In a second aspect, an electronic device is provided, which is configured to perform the method provided in the first aspect.
Optionally, the electronic device includes: smart phones, tablet computers, VR devices, smart glasses, smart televisions, or smart speakers.
The method provided by the application needs to perform 2 times of matrix product operation when the recurrent neural network performs operation calculation at the t-th layer, and then performs one time of matrix addition operation, and the technical scheme of the application firstly performs addition operation on 2 matrixes to obtain (X)t+St-1) Then carrying out matrix multiplication operation so as to reduce 1 time of matrix multiplication operation, reduce calculation quantity, raise calculation energy consumption of chip and reduce power, in addition, it can reduce (X)t+St-1) And the weight W is respectively determined as cyclic forwarding data and broadcast data and is realized through two ports, and the cyclic forwarding data and the broadcast data are transmitted through the two portsCompared with the prior art (for example, an H-type structure patent of the Council of China and Carniu), the data forwarding amount of one port is reduced, the data transmission amount of the port of the main processing circuit can be reduced, and the data forwarding amount of the conversion circuit is also reduced, in addition, the 4P6T switch is arranged, so that when the data are forwarded circularly, the port of the main processing circuit is more flexibly selected, and when a pin connected with the P port of the main processing circuit breaks down, the pin can be replaced through the 4P6T, the reliability of the chip is improved, the calculation efficiency is improved, the data forwarding amount and the calculation amount of the slave processing circuit are the same, the data forwarding is more balanced, the calculation efficiency is improved, and the user experience is improved.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a schematic diagram of the connection of an artificial intelligence chip according to the present invention.
Fig. 2 is a schematic flow chart of a speech processing method provided by the present invention.
Fig. 3 is a schematic diagram of the architecture of the recurrent neural network provided by the present invention.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.
The embodiments of the present application will be described below with reference to the drawings.
The term "and/or" in this application is only one kind of association relationship describing the associated object, and means that there may be three kinds of relationships, for example, a and/or B, which may mean: a exists alone, A and B exist simultaneously, and B exists alone. In addition, the character "/" in this document indicates that the former and latter related objects are in an "or" relationship.
The "plurality" appearing in the embodiments of the present application means two or more. The descriptions of the first, second, etc. appearing in the embodiments of the present application are only for illustrating and differentiating the objects, and do not represent the order or the particular limitation of the number of the devices in the embodiments of the present application, and do not constitute any limitation to the embodiments of the present application. The term "connect" in the embodiments of the present application refers to various connection manners, such as direct connection or indirect connection, to implement communication between devices, which is not limited in this embodiment of the present application.
In the present application, "|" means an absolute value.
Referring to fig. 1, fig. 1 provides a schematic diagram of an artificial intelligence chip structure, as shown in fig. 1, the artificial intelligence chip structure includes: a main processing circuit 101, 2 multi-way selection switches (4P6T), 6 groups of slave processing circuits, each group of slave processing circuits comprising: a plurality of slave processing circuits 102 and 1 forwarding circuit 103, the multiplexer switch being 4P 6T; the 6 ports of the main processing circuit are respectively connected with each forwarding circuit of the 6 groups of slave processing circuits, and each forwarding circuit is respectively connected with the broadcast ports of a plurality of slave processing circuits of the same group of slave processing circuits; the other 8 ports of the main processing circuit are respectively connected with 4P ports of 2 4P6T switches, and 6T ports of the 2 4P6T switches are respectively connected with a first slave processing circuit and a second slave processing circuit which are adjacent in each group of slave processing circuits; the slave processing circuit is also connected to adjacent other slave processing circuits within the same group of slave processing circuits via two forwarding ports.
Referring to fig. 2, fig. 2 also provides a speech processing method, where the method is performed by an electronic device, and the electronic device may include the artificial intelligence chip structure shown in fig. 1, and the method is shown in fig. 2, and includes the following steps:
step S201, the electronic equipment acquires a first voice, and the electronic equipment extracts input data X of the first voice at t-1 momentt-1Inputting the data Xt-1Determining the input data as the input data of the recurrent neural network at the t-1 th moment;
the electronic device may obtain the first voice through a microphone or other audio capture device. The input data may be extracted through an existing feature extraction network, and a specific extraction method is not limited herein.
Step S202, the electronic equipment inputs data Xt-1Inputting the weight value W into a main processing circuit, calling the weight value W of the recurrent neural network by the main processing circuit, and executing t-1 layer operation to obtain a hidden layer output result S at t-1 momentt-1And t-1 layer output result Ot-1；
Step S203, the main processing circuit receives the input data X of the first voice at the time tt(ii) a Mixing XtAnd St-1Performing an addition operation to obtain (X)t+St-1) (ii) a Will (X)t+St-1) Determining the data to be cyclic conversion data, determining the weight W to be broadcast forwarding data, cutting the broadcast forwarding data into a plurality of broadcast forwarding data blocks, respectively broadcasting the broadcast forwarding data blocks to a forwarding circuit through delta ports, cutting the cyclic forwarding data into alpha groups of cyclic forwarding data blocks, and sending the alpha groups of cyclic forwarding data blocks to a first slave processing circuit and a second slave processing circuit through a 4P6T switch;
step S204, the forwarding circuit forwards the received broadcast forwarding data block to a plurality of slave processing circuits in the same group of slave processing circuits; when the P port of the 4P6T switch receives a group of cyclic forwarding data blocks, one T port connecting one group of T ports (the port connected with the first slave processing circuit) is sent to the first slave processing circuit, and when the P port of the 4P6T switch receives another group of cyclic forwarding data blocks, the other T port connecting the other group of T ports (the T port connected with the second slave processing circuit) is sent to the second slave processing circuit;
step S205, when the first slave processing circuit receives a group of cyclic forwarding data blocks, intercepting a local cyclic forwarding data block from the group of cyclic forwarding data blocks, and forwarding the remaining cyclic forwarding data blocks to other slave processing circuits anticlockwise; when the second slave processing circuit receives another group of cyclic forwarding data blocks, intercepting local cyclic forwarding data blocks from the group of cyclic forwarding data blocks, and forwarding the rest cyclic forwarding data blocks to other slave processing circuits clockwise;
step S206, the slave processing circuit receives the residual cyclic forwarding data block through one forwarding port, receives the broadcast forwarding data block through the broadcast port, intercepts the local cyclic forwarding data block from the residual cyclic forwarding data block, and sends other cyclic forwarding data blocks to other adjacent slave processing circuits through another forwarding port; performing inner product operation on the local circulating forwarding data block and the broadcasting forwarding data block to obtain an operation result, and sending the operation result to a forwarding circuit through a broadcasting port;
step S207, the forwarding circuit forwards the operation result to the main processing circuit; the main processing circuit obtains the hidden layer output S of the t-th layer according to the operation resulttAnd the tth output result Ot(ii) a Basis of artificial intelligence chip St、OtAnd executing subsequent operation of the t-th layer to obtain a result of the recurrent neural network, and obtaining the meaning of the first voice according to the result of the recurrent neural network.
The artificial intelligence chip is based on St、OtThe subsequent operation of the t-th layer can be referred to the operation of the t-th layer to obtain the output result of the corresponding layer, and the meaning of the first voice obtained according to the output result can be confirmed by adopting the existing recurrent neural network, such as a recurrent neural network operation system of google.
The technical scheme provided by the application can reduce the calculation amount of the recurrent neural network, the recurrent neural network is a neural network model commonly used for speech translation, and the structure of the recurrent neural network is shown in figure 3 and comprises an input layer, a hidden layer and an output layer, wherein the output structure of the hidden layer is used as input data of the hidden layer at the next moment.
As shown in fig. 3, the output result of the hidden layer at time t is the output of the hidden layer at the next time t +1, for example.
As shown in FIG. 3, where W represents the weight, Xt-1Input data of the input layer representing the time t-1, XtInput data of the input layer representing time t, St-1Output result of hidden layer representing time t-1, Ot-1The output result of the output layer at the time t-1 is shown;
taking time t as an example:
St＝Xt×W+St-1×W
Ot＝f(St)
where f represents an activation function including, but not limited to: sigmoid function, tanh function, etc.
Of course, in practical applications, other activation functions may be used.
The main processing circuit may further include: and the activation module executes activation operation, and the activation operation can be specifically the operation of executing the activation function.
As shown in fig. 3, when the recurrent neural network performs operation calculation at the t-th layer, it needs to perform matrix multiplication operation 2 times, and then perform matrix addition operation one time, whereas the technical solution of the present application performs addition operation on 2 matrices to obtain (X)t+St-1) Then carrying out matrix multiplication operation so as to reduce 1 time of matrix multiplication operation, reduce calculation quantity, raise calculation energy consumption of chip and reduce power, in addition, it can reduce (X)t+St-1) And rightThe value W is respectively determined as cycle forwarding data and broadcast data and is realized through two ports, so that the forwarding data volume of one port is reduced compared with the broadcasting and cycle forwarding of one port, compared with the prior art (such as an H-type structure patent of Zhongkehan Ji), the data transmission quantity of the port of the main processing circuit can be reduced, and the forwarding data volume of the conversion circuit is also reduced, in addition, a 4P6T switch is arranged, so that the port selection of the main processing circuit is more flexible when the data is cyclically forwarded, and when a pin connected with the P port of the main processing circuit fails, the pin can be replaced through 4P6T, the reliability of a chip is improved, the calculation efficiency is improved, the forwarding data volume and the calculation volume of the slave processing circuits are the same, the data forwarding can be more balanced, and the calculation efficiency is improved, and the user experience is improved.
The intercepting of the local loop forwarding data block from the group of loop forwarding data blocks may specifically include: a row of element values or a column of element values is intercepted from a group of loop forwarding data blocks to determine the local loop forwarding data blocks. Can be (X)t+St-1) A row of element values or a column of element values.
The embodiment of the application also provides electronic equipment, and the electronic equipment is used for executing the method.
The electronic device includes: smart phones, tablet computers, VR devices, smart glasses, smart televisions, or smart speakers.
The foregoing detailed description of the embodiments of the present application has been presented to illustrate the principles and implementations of the present application, and the above description of the embodiments is only provided to help understand the method and the core concept of the present application; meanwhile, for a person skilled in the art, according to the idea of the present application, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present application.
Claims (4)
1. A speech processing method, characterized in that the method is performed by an electronic device comprising: an artificial intelligence chip, the artificial intelligence chip structure comprising: the main processing circuit, 2 multiplexer, 6 groups follow treatment circuit, every group follow treatment circuit includes: a plurality of slave processing circuits and 1 forwarding circuit, wherein the multiplexer switch is 4P 6T; the 6 ports of the main processing circuit are respectively connected with each forwarding circuit of the 6 groups of slave processing circuits, and each forwarding circuit is respectively connected with the broadcast ports of a plurality of slave processing circuits of the same group of slave processing circuits; the other 8 ports of the main processing circuit are respectively connected with 4P ports of 2 4P6T switches, and 6T ports of the 2 4P6T switches are respectively connected with a first slave processing circuit and a second slave processing circuit which are adjacent in each group of slave processing circuits; the slave processing circuit is also connected with other adjacent slave processing circuits in the same group of slave processing circuits through two forwarding ports; the method comprises the following steps:
the electronic equipment acquires a first voice, and the electronic equipment extracts input data X of the first voice at the t-1 momentt-1Inputting the data Xt-1Determining the input data as the input data of the recurrent neural network at the t-1 th moment;
the electronic equipment inputs data Xt-1Inputting the weight value W into a main processing circuit, calling the weight value W of the recurrent neural network by the main processing circuit, and executing t-1 layer operation to obtain a hidden layer output result S at t-1 momentt-1And t-1 layer output result Ot-1；
The main processing circuit receives the input data X of the first voice at the time tt(ii) a Mixing XtAnd St-1Performing an addition operation to obtain (X)t+St-1) (ii) a Will (X)t+St-1) Determining the data to be cyclic conversion data, determining the weight W to be broadcast forwarding data, cutting the broadcast forwarding data into a plurality of broadcast forwarding data blocks, respectively broadcasting the broadcast forwarding data blocks to a forwarding circuit through delta ports, cutting the cyclic forwarding data into alpha groups of cyclic forwarding data blocks, and sending the alpha groups of cyclic forwarding data blocks to a first slave processing circuit and a second slave processing circuit through a 4P6T switch;
the forwarding circuit forwards the received broadcast forwarding data block to a plurality of slave processing circuits in the same group of slave processing circuits; when the P port of the 4P6T switch receives a group of cyclic forwarding data blocks, one T port connected with one group of T ports is sent to the first slave processing circuit, and when the P port of the 4P6T switch receives another group of cyclic forwarding data blocks, the other T port connected with the other group of T ports is sent to the second slave processing circuit;
when the first slave processing circuit receives a group of cyclic forwarding data blocks, intercepting a local cyclic forwarding data block from the group of cyclic forwarding data blocks, and forwarding the rest cyclic forwarding data blocks to other slave processing circuits anticlockwise; when the second slave processing circuit receives another group of cyclic forwarding data blocks, intercepting local cyclic forwarding data blocks from the group of cyclic forwarding data blocks, and forwarding the rest cyclic forwarding data blocks to other slave processing circuits clockwise;
the slave processing circuit receives the residual cyclic forwarding data block through one forwarding port, receives the broadcast forwarding data block through the broadcast port, intercepts the local cyclic forwarding data block from the residual cyclic forwarding data block, and sends other cyclic forwarding data blocks to other adjacent slave processing circuits through another forwarding port; performing inner product operation on the local circulating forwarding data block and the broadcasting forwarding data block to obtain an operation result, and sending the operation result to a forwarding circuit through a broadcasting port;
the forwarding circuit forwards the operation result to the main processing circuit; the main processing circuit obtains the hidden layer output S of the t-th layer according to the operation resulttAnd the tth output result Ot；
Basis of artificial intelligence chip St、OtAnd executing subsequent operation of the t-th layer to obtain a result of the recurrent neural network, and obtaining the meaning of the first voice according to the result of the recurrent neural network.
2. The method of claim 1, wherein the main processing circuit further comprises: an activation module that performs an activation operation.
3. An electronic device, characterized in that the electronic device is adapted to perform the method of any of claims 1-2.
4. The electronic device of claim 3,
the electronic device includes: smart phones, tablet computers, VR devices, smart glasses, smart televisions, or smart speakers.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011183800.3A CN112289310A (en) | 2020-10-29 | 2020-10-29 | Voice processing method and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011183800.3A CN112289310A (en) | 2020-10-29 | 2020-10-29 | Voice processing method and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112289310A true CN112289310A (en) | 2021-01-29 |
Family
ID=74353474
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011183800.3A Withdrawn CN112289310A (en) | 2020-10-29 | 2020-10-29 | Voice processing method and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112289310A (en) |
-
2020
- 2020-10-29 CN CN202011183800.3A patent/CN112289310A/en not_active Withdrawn
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20210117810A1 (en) | On-chip code breakpoint debugging method, on-chip processor, and chip breakpoint debugging system | |
| CN110263909B (en) | Image recognition method and device | |
| CN109522052B (en) | Computing device and board card | |
| KR102325602B1 (en) | Device and method to process data in parallel | |
| CN108932124A (en) | neural network model compression method, device, terminal device and storage medium | |
| JP2020135872A (en) | Object recognition method, device, and network | |
| CN109726822B (en) | Operation method, device and related product | |
| CN111524516A (en) | Control method based on voice interaction, server and display device | |
| CN110909870B (en) | Training device and method | |
| CN111353591A (en) | Computing device and related product | |
| CN106528490B (en) | FPGA heterogeneous acceleration computing device and system | |
| TW202145079A (en) | Operation execution method and device, electronic equipment and storage medium | |
| CN115766159A (en) | Private data processing method and device and electronic equipment | |
| CN109711540B (en) | Computing device and board card | |
| CN110059809B (en) | Computing device and related product | |
| CN111160542A (en) | Integrated circuit chip device and related product | |
| CN112289310A (en) | Voice processing method and electronic equipment | |
| CN111814972B (en) | Neural network convolution operation acceleration method based on FPGA | |
| CN112491852A (en) | Signature processing method for electronic commerce and electronic equipment | |
| CN112309376A (en) | Telemedicine voice processing method and electronic equipment | |
| CN109711538B (en) | Operation method, device and related product | |
| WO2023124371A1 (en) | Data processing apparatus and method, and chip, computer device and storage medium | |
| CN109740729B (en) | Operation method, device and related product | |
| CN109740730B (en) | Operation method, device and related product | |
| CN112491555B (en) | Medical electronic signature processing method and electronic equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20210527 Address after: 2 / F, building 5, Changyuan new material port engineering technology center, Gaoxin Middle Road, Yuehai street, Nanshan District, Shenzhen, Guangdong 518000 Applicant after: Shenzhen Guangcheng Jierui Technology Co.,Ltd. Address before: 21-3, building 10, 26 Nanbin Road, Nan'an District, Chongqing 400000 Applicant before: Chongqing seamless splicing Intelligent Technology Co.,Ltd. |
|
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210129 |