WO2017157093A1 - Input method, device, and computer storage medium - Google Patents

Input method, device, and computer storage medium Download PDF

Info

Publication number
WO2017157093A1
WO2017157093A1 PCT/CN2017/070905 CN2017070905W WO2017157093A1 WO 2017157093 A1 WO2017157093 A1 WO 2017157093A1 CN 2017070905 W CN2017070905 W CN 2017070905W WO 2017157093 A1 WO2017157093 A1 WO 2017157093A1
Authority
WO
WIPO (PCT)
Prior art keywords
word
text
words
terminal
vector
Prior art date
Application number
PCT/CN2017/070905
Other languages
French (fr)
Chinese (zh)
Inventor
许林
Original Assignee
中兴通讯股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 中兴通讯股份有限公司 filed Critical 中兴通讯股份有限公司
Publication of WO2017157093A1 publication Critical patent/WO2017157093A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/16—Sound input; Sound output
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F3/00—Input arrangements for transferring data to be processed into a form capable of being handled by the computer; Output arrangements for transferring data from processing unit to output unit, e.g. interface arrangements
    • G06F3/01—Input arrangements or combined input and output arrangements for interaction between user and computer
    • G06F3/02—Input arrangements using manually operated switches, e.g. using keyboards or dials
    • G06F3/023—Arrangements for converting discrete items of information into a coded form, e.g. arrangements for interpreting keyboard generated codes as alphanumeric codes, operand codes or instruction codes
    • G06F3/0233—Character input methods
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/26—Speech to text systems

Definitions

  • the present invention relates to terminal technologies, and more particularly to a method, device and computer storage medium for implementing input.
  • the input vocabulary contact function is provided to reduce the number of unnecessary keyboard strokes.
  • the traditional keyboard input method's contact prompt function is based on the holistic nature of the words. When typing the word “North”, it will prompt the words “Beijing”, “Northern Side” and “Northern”. These prompt words are all “North”. The word at the beginning of the word.
  • the traditional keyboard input method's contact prompt function is based on the word integrity. When the complete word is input, it has no prompt function, and the user needs to input a new word for association and prompt.
  • the embodiment of the invention provides a method, a device and a computer storage medium for realizing input, which can realize association of input words.
  • An embodiment of the present invention provides a method for implementing an input prompt, including:
  • the terminal collects a voice signal, and converts the collected voice signal into text text;
  • the terminal extracts words in the text text
  • the terminal acquires and displays the words associated with the words in the text text.
  • the terminal after the terminal extracts the words in the text text, and before the terminal acquires the words associated with the words in the text text and displays the words, the terminal further includes:
  • the terminal determines the part of speech of the word in the text text
  • the terminal acquires a word similar to the word in the text text and displays:
  • the terminal acquires and displays a word associated with a word other than the virtual word in the text text.
  • the terminal extracting words in the text text includes:
  • the terminal divides the text text into words in units of words to obtain words in the text text.
  • the terminal divides the text text into units of words to obtain words in the text text, including:
  • the terminal determines a division scheme, and divides the text text according to the determined division scheme to obtain words in the text text.
  • the determining, by the terminal, the dividing scheme includes:
  • the terminal calculates a state vector corresponding to each word in the text text, calculates a state transition matrix, calculates a score of each word segmentation scheme according to a state vector corresponding to all words and a state transition matrix, and determines a word segmentation scheme with the largest score as described Division plan.
  • the terminal acquiring a word associated with the word in the text text and displaying includes:
  • the terminal acquires N words that are the closest to each word in the text text or each word except the virtual word; wherein N is an integer greater than or equal to 1;
  • the terminal calculates a score of the obtained M ⁇ N words;
  • M is the number of words in the text text or the number of words other than the virtual word;
  • the terminals are displayed in descending order of score.
  • the terminal acquires N words that are the closest to each word in the text text or each word except the function word, including:
  • the terminal converts each word in the text text or each word except the virtual word into a first word vector, and acquires a word corresponding to the N second word vectors having the smallest distance from each of the first word vectors.
  • the terminal converting each word in the text text or each word except the virtual word into the first word vector includes:
  • the acquiring a word corresponding to the N second word vectors having the smallest distance from each of the first word vectors includes:
  • the terminal calculates a distance between each of the first word vectors and each of the second word vectors in the preset word vector space, and acquires words corresponding to the N second vectors that minimize the calculated distance.
  • the terminal calculates a distance between each first word vector and each second word vector in a preset word vector space, including:
  • the terminal according to the formula Calculating a distance between the first word vector and the second word vector
  • p ; p (0, + ⁇ );
  • d represents the distance between the first word vector and the second word vector
  • a is the first word vector
  • b is the second word vector
  • d represents the first The Manhattan distance between the word vector and the second word vector
  • p ⁇ d represents the Chebyshev distance between the first word vector and the second word vector.
  • the terminal calculates the scores of the obtained M ⁇ N words including:
  • c is the score of each of the M ⁇ N words
  • w is the weight of the word corresponding to the first word vector corresponding to each of the M ⁇ N words
  • L is the first word The distance between the vector and the second word vector corresponding to each of the M x N words.
  • the weight of the word corresponding to the first word vector is Where s is the position of the word corresponding to the first word vector in the text text; and S represents a word corresponding to the first word vector after all words in the text character or words other than the virtual word are numbered The number.
  • An embodiment of the present invention further provides a method for implementing an input prompt, including:
  • the server receives the voice signal collected from the terminal, and converts the received voice signal into text text;
  • the server obtains a word in the text text
  • the server obtains the words associated with the words in the text text and sends them to the terminal.
  • the server after the server obtains the word in the text text, before the server acquires the word associated with the word in the text text and sends the word to the terminal, the server further includes:
  • the server determines the part of speech of the word in the text text
  • the server acquires a word associated with a word other than the virtual word in the text text and transmits the word to the terminal.
  • An embodiment of the present invention further provides a method for implementing an input prompt, including:
  • the terminal collects a voice signal, and sends the collected voice signal to the server;
  • the terminal receives and displays a word associated with the word in the text text corresponding to the voice signal sent by the server.
  • the embodiment of the invention further provides a terminal, including:
  • the first collection module is configured to collect a voice signal
  • a first conversion module configured to convert the collected voice signal into text text
  • a first processing module configured to extract words in the textual text; to acquire and display words associated with words in the textual text.
  • a first receiving module configured to receive a voice signal collected from the terminal
  • a second processing module configured to acquire a word in the text text; acquire a word associated with a word in the text text;
  • a second sending module configured to send the collected voice signal to the server
  • the second receiving module is configured to receive and display a word associated with the word in the text text corresponding to the voice signal sent by the server.
  • FIG. 3 is a flowchart of another method (terminal) for implementing an input prompt according to an embodiment of the present invention
  • FIG. 4 is a schematic structural diagram of a terminal according to an embodiment of the present invention.
  • FIG. 6 is a schematic structural diagram of another terminal according to an embodiment of the present invention.
  • Step 100 The terminal collects a voice signal, and converts the collected voice signal into text text.
  • the terminal can collect the voice analog signal, convert the voice analog signal into a voice digital signal, filter the voice digital signal, and convert the filtered voice digital signal into text text, which can be specifically known by those skilled in the art.
  • the implementation is not intended to limit the scope of the present invention, and details are not described herein.
  • Step 101 The terminal extracts words in the text text.
  • the terminal divides the text text into words in units of words to obtain words in the text.
  • the terminal calculates a state vector corresponding to each word in the text text, calculates a state transition matrix, calculates a score of each word segmentation scheme according to a state vector corresponding to all words and a state transition matrix, and determines a word segmentation scheme with the largest score as a division scheme.
  • each element in the state vector represents the probability that the word is the state corresponding to the element.
  • the terminal converts each word in the text into a word vector, and calculates a state vector of the word based on the word vector.
  • the terminal can calculate the state vector of the word according to the word vector using a multi-layer neural network model.
  • the state corresponding to a word can be B, or E, or I, or S, where B represents At the beginning of the word, E is the end of the word, I is the middle of the word, and S is a word.
  • the state vector can be expressed as (P B , P E , P I , P S ), where P B represents the probability that the word is the beginning of the word, P E represents the probability that the word is the end of the word, P I represents the probability The word is the probability in the middle of the word, and P S is the probability that the word is a word.
  • s i is the score of the i-th word segmentation scheme
  • a ij is the probability that the state of the (i-1)th word of the i-th participle program shifts to the state of the j-th word
  • f ij is the i-th word-dividing scheme The state of the jth word.
  • Step 102 The terminal acquires a word associated with a word in the text text and displays the word.
  • step 101 further includes:
  • the terminal determines the part of speech of the word in the text:
  • the terminal calculates a state vector corresponding to each word in the text text, calculates a state transition matrix, calculates a score of each program for determining the part of speech according to the state vector and the state transition matrix corresponding to all words, and determines according to the scheme of determining the part of speech with the largest score.
  • the part of speech of each word in the text is a state vector corresponding to each word in the text text.
  • the specific implementation process described above and the terminal divide the text text into words in units of words to obtain similar words in the text text, which can be specifically implemented by referring to the above implementation process.
  • step 102 the terminal acquires a word associated with the word in the text text and displays:
  • the N words that the terminal obtains the most similar to each word in the text or each word except the virtual word include:
  • the terminal converts each word in the text text or each word except the function word into a first word vector, and acquires a word corresponding to the N second word vectors having the smallest distance from each of the first word vectors.
  • the terminal searches for each word in the text text or the first word vector corresponding to each word except the virtual word in the second correspondence between the preset word and the word vector.
  • the terminal calculates a distance between each of the first word vectors and each of the second word vectors in the preset word vector space, and acquires words corresponding to the N second vectors that minimize the calculated distance.
  • the word vector is used to represent the word, and the first word of the word that is similar to the text is not necessarily the same as the word input by the user, so that the associated word is more abundant.
  • ⁇ is an angle between the first word vector and the second word vector for indicating a distance between the first word vector and the second word vector
  • a is a first word vector
  • b is the second word vector.
  • ⁇ ranges from 0 to ⁇ ;
  • d is the distance between the first word vector and the second word vector
  • a is the first word vector
  • b is the second word vector
  • d represents the Chebyshev distance between the first word vector and the second word vector.
  • the scores of the M ⁇ N words obtained by the terminal calculation include:
  • the terminal calculates the score of each of the M ⁇ N words, and adds the sum of the scores corresponding to the words that appear multiple times in the M ⁇ N words as the score of the word; appears once in the M ⁇ N words The word corresponds to the score as the score of the word.
  • the terminal calculates the score of each of the M ⁇ N words including:
  • the weight of the word corresponding to the first word vector is Where s is the position of the word corresponding to the first word vector in the text text.
  • the position of the word corresponding to the first word vector in the text text refers to the number of the word corresponding to the first word vector after all words in the text text or words other than the virtual word are numbered from beginning to end.
  • the text is "He loves to play” and is divided into three words: he, love, play, number “he” as 1, “love” number 2, "play” number 3; or "love” number For 1, the "playing" number is 2.
  • Step 200 The server receives the voice signal collected by the terminal, and converts the received voice signal into text text.
  • Step 201 The server acquires a word in the text text.
  • step 101 The specific implementation of this step is the same as step 101.
  • Step 202 The server acquires a word associated with the word in the text text and sends the word to the terminal.
  • the server acquires a word similar to the word in the text text and the terminal in step 102 acquires the word similar to the word in the text text.
  • the server may also send a score of a word similar to the word in the text text to the terminal.
  • the server sorts the scores of the words that are similar to the part of speech in the text text from high to low, and sends the sorted words that are similar to the words in the text text to the terminal.
  • the server determines the part of speech of the word in the text text; accordingly,
  • the server obtains words similar to the words in the text and sends them to the terminal including:
  • the server acquires a word similar to a word other than the virtual word in the text text and transmits the word to the terminal.
  • the server determines that the part of speech of the word in the text text is implemented in the same manner as the word form of the word in the terminal determining the text word.
  • an embodiment of the present invention further provides a method for implementing an input prompt, including:
  • Step 300 The terminal collects a voice signal, and sends the collected voice signal to the server.
  • Step 301 The terminal receives and displays a word associated with a word in a text text corresponding to the voice signal sent by the server.
  • the terminal can also display the scores of words that are close to the words in the text text from high to low.
  • an embodiment of the present invention further provides a terminal, including:
  • the first collection module is configured to collect a voice signal
  • a first conversion module configured to convert the collected voice signal into text text
  • the first processing module is configured to acquire words in the text text; acquire words that are similar to the words in the text text and display them.
  • the first processing module is configured to:
  • the first acquisition module, the first conversion module, and the first processing module in the terminal may be used by a central processing unit (CPU) and a digital signal in the terminal in an actual application.
  • CPU central processing unit
  • a DSP Digital Signal Processor
  • MCU Microcontroller Unit
  • FPGA Field-Programmable Gate Array
  • an embodiment of the present invention further provides a server, including:
  • a first receiving module configured to receive a voice signal collected from the terminal
  • a second conversion module configured to convert the received voice signal into text text
  • a second processing module configured to obtain a word in the text text; to obtain a word that is similar to the word in the text text;
  • the second processing module is configured to:
  • the second conversion module and the second processing module in the server may be implemented by a CPU, a DSP, an MCU or an FPGA in the server in an actual application; a first receiving module in the server And the first sending module, in practical applications, can pass the communication module (including: basic communication suite, operating system, communication module, standardized interface and protocol, etc.) Transceiver antenna implementation.
  • an embodiment of the present invention further provides a terminal, including:
  • a second collection module configured to collect a voice signal
  • a second sending module configured to send the collected voice signal to the server
  • the second receiving module is configured to receive and display a word similar to the word in the text text corresponding to the voice signal from the server.
  • the second collection module in the terminal may be implemented by a CPU, a DSP, an MCU, or an FPGA in the terminal in a practical application; a second sending module and a second receiving module in the terminal, In practical applications, it can be realized through communication modules (including: basic communication suite, operating system, communication module, standardized interface and protocol, etc.) and transceiver antennas.
  • communication modules including: basic communication suite, operating system, communication module, standardized interface and protocol, etc.
  • the disclosed apparatus and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner such as: multiple units or components may be combined, or Can be integrated into another system, or some features can be ignored or not executed.
  • the coupling, or direct coupling, or communication connection of the components shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may be electrical, mechanical or other forms. of.
  • the units described above as separate components may or may not be physically separated, and the components displayed as the unit may or may not be physical units, that is, may be located in one place or distributed to multiple network units; Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • the foregoing may be completed by a program instruction related hardware, where the foregoing program may be stored in a computer readable storage medium, and when executed, the program includes the steps of the foregoing method embodiment; and the foregoing storage medium includes: mobile storage A device that can store program code, such as a device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
  • ROM read-only memory
  • RAM random access memory
  • magnetic disk or an optical disk.
  • the above-described integrated unit of the present invention may be stored in a computer readable storage medium if it is implemented in the form of a software function module and sold or used as a standalone product.
  • the technical solution of the embodiments of the present invention may be embodied in the form of a software product in essence or in the form of a software product stored in a storage medium, including a plurality of instructions.
  • a computer device (which may be a personal computer, server, or network device, etc.) is caused to perform all or part of the methods described in various embodiments of the present invention.
  • the foregoing storage medium includes various media that can store program codes, such as a mobile storage device, a ROM, a RAM, a magnetic disk, or an optical disk.
  • the technical solution of the embodiment of the present invention collects a voice signal by using a terminal, converts the collected voice signal into text text, extracts a word in the text text, acquires a word associated with the word in the text text, and displays the The association of the words entered.

Abstract

An input method, device, and computer storage medium. The input method comprises: a terminal collects a voice signal, and converts the collected voice signal into a text (100); the terminal extracts a word from the text (101); and the terminal acquires another word related to the word in the text and displays the same (102).

Description

一种实现输入的方法、装置及计算机存储介质Method, device and computer storage medium for realizing input 技术领域Technical field
本发明涉及终端技术，尤指一种实现输入的方法、装置及计算机存储介质。The present invention relates to terminal technologies, and more particularly to a method, device and computer storage medium for implementing input.
背景技术Background technique
传统的键盘输入时，会提供输入词汇的联系功能，以减少不必要的键盘敲击次数。传统的键盘输入法的联系提示功能是基于词语的整体性，必然当键入“北”字时，会提示“北京”，“北边”和“北方”等词语，这些提示词语都是以“北”字开头的词。When the traditional keyboard is input, the input vocabulary contact function is provided to reduce the number of unnecessary keyboard strokes. The traditional keyboard input method's contact prompt function is based on the holistic nature of the words. When typing the word "North", it will prompt the words "Beijing", "Northern Side" and "Northern". These prompt words are all "North". The word at the beginning of the word.
传统的键盘输入法的联系提示功能是基于词语整体性的，当输入完整词语后就无提示功能，需要用户输入新的一个字以进行联想和提示。The traditional keyboard input method's contact prompt function is based on the word integrity. When the complete word is input, it has no prompt function, and the user needs to input a new word for association and prompt.
发明内容Summary of the invention
本发明实施例提出了一种实现输入的方法、装置及计算机存储介质，能够实现对输入的词语的联想。The embodiment of the invention provides a method, a device and a computer storage medium for realizing input, which can realize association of input words.
本发明实施例提出了一种实现输入提示的方法，包括：An embodiment of the present invention provides a method for implementing an input prompt, including:
终端采集语音信号，将采集的语音信号转换为文本文字；The terminal collects a voice signal, and converts the collected voice signal into text text;
终端提取所述文本文字中的词；The terminal extracts words in the text text;
终端获取与所述文本文字中的词相关联的词并显示。The terminal acquires and displays the words associated with the words in the text text.
作为一种实施方式，所述终端提取所述文本文字中的词后，在所述终端获取与所述文本文字中的词相关联的词并显示之前，还包括：As an implementation manner, after the terminal extracts the words in the text text, and before the terminal acquires the words associated with the words in the text text and displays the words, the terminal further includes:
所述终端确定所述文本文字中的词的词性；The terminal determines the part of speech of the word in the text text;
所述终端获取与文本文字中的词相近的词并显示包括：The terminal acquires a word similar to the word in the text text and displays:
所述终端获取与所述文本文字中除虚词以外的词相关联的词并显示。 The terminal acquires and displays a word associated with a word other than the virtual word in the text text.
作为一种实施方式，所述终端提取所述文本文字中的词，包括：As an implementation manner, the terminal extracting words in the text text includes:
所述终端将所述文本文字以词为单位进行划分得到所述文本文字中的词。The terminal divides the text text into words in units of words to obtain words in the text text.
作为一种实施方式，所述终端将文本文字以词为单位进行划分得到文本文字中的词包括：As an implementation manner, the terminal divides the text text into units of words to obtain words in the text text, including:
所述终端确定划分方案，按照确定的划分方案将所述文本文字进行划分得到所述文本文字中的词。The terminal determines a division scheme, and divides the text text according to the determined division scheme to obtain words in the text text.
作为一种实施方式，所述终端确定划分方案包括：As an implementation manner, the determining, by the terminal, the dividing scheme includes:
所述终端计算文本文字中的每一个字对应的状态向量，计算状态转移矩阵，根据所有字对应的状态向量和状态转移矩阵计算每一种分词方案的分数，确定分数最大的分词方案为所述划分方案。The terminal calculates a state vector corresponding to each word in the text text, calculates a state transition matrix, calculates a score of each word segmentation scheme according to a state vector corresponding to all words and a state transition matrix, and determines a word segmentation scheme with the largest score as described Division plan.
作为一种实施方式，所述终端获取与所述文本文字中的词相关联的词并显示包括：As an implementation manner, the terminal acquiring a word associated with the word in the text text and displaying includes:
所述终端获取与所述文本文字中的每一个词或除虚词以外的每一个词最相近的N个词；其中，N为大于或等于1的整数；The terminal acquires N words that are the closest to each word in the text text or each word except the virtual word; wherein N is an integer greater than or equal to 1;
所述终端计算获得的M×N个词的得分；M为所述文本文字中的词的个数或除虚词以外的词的个数；The terminal calculates a score of the obtained M×N words; M is the number of words in the text text or the number of words other than the virtual word;
所述终端按照得分从高到低的顺序进行显示。The terminals are displayed in descending order of score.
作为一种实施方式，所述终端获取与文本文字中的每一个词或除虚词以外的每一个词最相近的N个词包括：As an implementation manner, the terminal acquires N words that are the closest to each word in the text text or each word except the function word, including:
所述终端将所述文本文字中的每一个词或除虚词以外的每一个词转换成第一词向量，获取与每一个第一词向量距离最小的N个第二词向量对应的词。The terminal converts each word in the text text or each word except the virtual word into a first word vector, and acquires a word corresponding to the N second word vectors having the smallest distance from each of the first word vectors.
作为一种实施方式，所述终端将文本文字中的每一个词或除虚词以外的每一个词转换成第一词向量包括：As an implementation manner, the terminal converting each word in the text text or each word except the virtual word into the first word vector includes:
所述终端在预先设置的词和词向量之间的第二对应关系中，查找所述 文本文字中的每一个词或所述除虚词以外的每一个词对应的第一词向量。The terminal searches for the second correspondence in a pre-set word and word vector Each word in the text or a first word vector corresponding to each word except the word.
作为一种实施方式，所述获取与每一个第一词向量距离最小的N个第二词向量对应的词包括：As an implementation manner, the acquiring a word corresponding to the N second word vectors having the smallest distance from each of the first word vectors includes:
所述终端计算每一个所述第一词向量与预先设置的词向量空间中的每一个第二词向量之间的距离，获取将计算得到的距离最小的N个第二向量对应的词。The terminal calculates a distance between each of the first word vectors and each of the second word vectors in the preset word vector space, and acquires words corresponding to the N second vectors that minimize the calculated distance.
作为一种实施方式，所述终端计算每一个第一词向量与预先设置的词向量空间中的每一个第二词向量之间的距离包括：As an implementation manner, the terminal calculates a distance between each first word vector and each second word vector in a preset word vector space, including:
所述终端按照公式
Figure PCTCN2017070905-appb-000001
计算所述第一词向量和所述第二词向量之间的距离；The terminal according to the formula
Figure PCTCN2017070905-appb-000001
Calculating a distance between the first word vector and the second word vector;
其中，θ为所述第一词向量和所述第二词向量之间的夹角，用于表示所述第一词向量和所述第二词向量之间的距离，a为所述第一词向量，b为所述第二词向量；Where θ is an angle between the first word vector and the second word vector, and is used to represent a distance between the first word vector and the second word vector, where a is the first a word vector, b is the second word vector;
或者，所述终端按照公式d＝||a-b||p计算所述第一词向量和所述第二词向量之间的距离；p＝(0,+∞]；Alternatively, the terminal calculates a distance between the first word vector and the second word vector according to the formula d=||ab|| p ; p=(0, +∞);
其中，d表示所述第一词向量和所述第二词向量之间的距离，a为所述第一词向量，b为所述第二词向量；当p＝1时，d表示第一词向量和所述第二词向量之间的曼哈顿距离，当p＝2时，d表示第一词向量和所述第二词向量之间的欧式距离。当p→∞时，d表示第一词向量和所述第二词向量之间的切比雪夫距离。Where d represents the distance between the first word vector and the second word vector, a is the first word vector, b is the second word vector; when p=1, d represents the first The Manhattan distance between the word vector and the second word vector, when p=2, d represents the Euclidean distance between the first word vector and the second word vector. When p→∞, d represents the Chebyshev distance between the first word vector and the second word vector.
作为一种实施方式，所述终端计算获得的M×N个词的得分包括：As an implementation manner, the terminal calculates the scores of the obtained M×N words including:
所述终端计算所述M×N个词中每一个词的分数，将所述M×N个词中出现多次的词对应的分数相加得到的和值作为该词的得分；将所述M×N个词中出现一次的词对应的分数作为该词的得分。The terminal calculates a score of each of the M×N words, and adds a sum value obtained by adding the scores corresponding to the plurality of words in the M×N words as a score of the word; The score corresponding to the word that appears once in M×N words is taken as the score of the word.
作为一种实施方式，所述终端计算M×N个词中每一个词的分数包括： As an implementation manner, the terminal calculates a score of each of the M×N words including:
所述终端按照公式c＝wL计算所述M×N个词中每一个词的分数；The terminal calculates a score of each of the M×N words according to a formula c=wL;
其中，c为所述M×N个词中每一个词的分数，w为M×N个词中每一个词对应的所述第一词向量对应的词的权重，L为所述第一词向量和M×N个词中每一个词对应的所述第二词向量之间的距离。Where c is the score of each of the M×N words, w is the weight of the word corresponding to the first word vector corresponding to each of the M×N words, and L is the first word The distance between the vector and the second word vector corresponding to each of the M x N words.
作为一种实施方式，所述第一词向量对应的词的权值为
Figure PCTCN2017070905-appb-000002
其中，s为所述第一词向量对应的词在所述文本文字中的位置；所述S表示文本文字中所有词或除虚词以外的词进行编号后，所述第一词向量对应的词的编号。As an implementation manner, the weight of the word corresponding to the first word vector is
Figure PCTCN2017070905-appb-000002
Where s is the position of the word corresponding to the first word vector in the text text; and S represents a word corresponding to the first word vector after all words in the text character or words other than the virtual word are numbered The number.
本发明实施例还提出了一种实现输入提示的方法，包括：An embodiment of the present invention further provides a method for implementing an input prompt, including:
服务器接收到来自终端采集的语音信号，将接收到的语音信号转换为文本文字；The server receives the voice signal collected from the terminal, and converts the received voice signal into text text;
服务器获取所述文本文字中的词；The server obtains a word in the text text;
服务器获取与所述文本文字中的词相关联的词并发送给终端。The server obtains the words associated with the words in the text text and sends them to the terminal.
作为一种实施方式，所述服务器获取所述文本文字中的词后，在所述服务器获取与所述文本文字中的词相关联的词并发送给终端之前还包括：As an implementation manner, after the server obtains the word in the text text, before the server acquires the word associated with the word in the text text and sends the word to the terminal, the server further includes:
所述服务器确定所述文本文字中的词的词性；The server determines the part of speech of the word in the text text;
所述服务器获取与文本文字中的词相近的词并发送给终端包括：The server acquires a word similar to the word in the text text and sends the word to the terminal, including:
所述服务器获取与所述文本文字中除虚词以外的词相关联的词并发送给所述终端。The server acquires a word associated with a word other than the virtual word in the text text and transmits the word to the terminal.
本发明实施例还提出了一种实现输入提示的方法，包括：An embodiment of the present invention further provides a method for implementing an input prompt, including:
终端采集语音信号，将采集的语音信号发送给服务器；The terminal collects a voice signal, and sends the collected voice signal to the server;
终端接收到所述服务器发送的与所述语音信号对应的文本文字中的词相关联的词并显示。The terminal receives and displays a word associated with the word in the text text corresponding to the voice signal sent by the server.
本发明实施例还提出了一种终端，包括：The embodiment of the invention further provides a terminal, including:
第一采集模块，配置为采集语音信号； The first collection module is configured to collect a voice signal;
第一转换模块，配置为将采集的语音信号转换为文本文字；a first conversion module configured to convert the collected voice signal into text text;
第一处理模块，配置为提取所述文本文字中的词；获取与所述文本文字中的词相关联的词并显示。a first processing module configured to extract words in the textual text; to acquire and display words associated with words in the textual text.
作为一种实施方式，所述第一处理模块配置为：As an implementation manner, the first processing module is configured to:
获取文本文字中的词；确定所述文本文字中的词的词性；获取与所述文本文字中除虚词以外的词相关联的词并显示。Obtaining words in the text text; determining the part of speech of the words in the text text; acquiring words associated with words other than the virtual words in the text text and displaying them.
本发明实施例还提出了一种服务器，包括：The embodiment of the invention further provides a server, including:
第一接收模块，配置为收到来自终端采集的语音信号；a first receiving module configured to receive a voice signal collected from the terminal;
第二转换模块，配置为将接收到的语音信号转换为文本文字；a second conversion module configured to convert the received voice signal into text text;
第二处理模块，配置为获取所述文本文字中的词；获取与所述文本文字中的词相关联的词；a second processing module configured to acquire a word in the text text; acquire a word associated with a word in the text text;
第一发送模块，配置为将与所述文本文字中的词相关联的词发送给终端。The first sending module is configured to send a word associated with the word in the text text to the terminal.
作为一种实施方式，所述第二处理模块配置为：As an implementation manner, the second processing module is configured to:
获取文本文字中的词；确定所述文本文字中的词的词性；获取与所述文本文字中除虚词以外的词相近的词。Obtaining words in the text text; determining the part of speech of the words in the text text; and acquiring words similar to words other than the virtual words in the text text.
本发明实施例还提出了一种终端，包括：The embodiment of the invention further provides a terminal, including:
第二采集模块，配置为采集语音信号；a second collection module configured to collect a voice signal;
第二发送模块，配置为将采集的语音信号发送给服务器；a second sending module, configured to send the collected voice signal to the server;
第二接收模块，配置为接收到所述服务器发送的与所述语音信号对应的文本文字中的词相关联的词并显示。The second receiving module is configured to receive and display a word associated with the word in the text text corresponding to the voice signal sent by the server.
本发明实施例还提出了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行本发明实施例所述的第一种应用于终端中的实现输入提示的方法。The embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the first application in the terminal according to the embodiment of the present invention. A way to implement input prompts.
本发明实施例还提出了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行本发明实施例 所述的第二种应用于服务器中的实现输入提示的方法。The embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the embodiment of the present invention. The second method is applied to a method of implementing input prompts in a server.
本发明实施例还提出了一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行本发明实施例所述的第二种应用于终端中的实现输入提示的方法。The embodiment of the present invention further provides a computer storage medium, where the computer storage medium stores computer executable instructions, and the computer executable instructions are used to execute the second application in the terminal according to the embodiment of the present invention. A way to implement input prompts.
与相关技术相比，本发明实施例的技术方案包括：终端采集语音信号，将采集的语音信号转换为文本文字；终端提取所述文本文字中的词；终端获取与所述文本文字中的词相关联的词并显示。通过本发明实施例的方案，获取与文本文字中的词相近的词并显示，实现了对输入的词语的联想。Compared with the related art, the technical solution of the embodiment of the present invention includes: the terminal collects the voice signal, converts the collected voice signal into text text; the terminal extracts the word in the text text; and the terminal acquires the word in the text text Associated words are displayed. Through the solution of the embodiment of the present invention, words similar to words in the text text are obtained and displayed, and association with the input words is realized.
附图说明DRAWINGS
下面对本发明实施例中的附图进行说明，实施例中的附图是用于对本发明的进一步理解，与说明书一起用于解释本发明，并不构成对本发明保护范围的限制。The drawings in the following description of the embodiments of the present invention are intended to illustrate the invention, and are not intended to limit the scope of the invention.
图1为本发明实施例实现输入提示的方法的流程图；FIG. 1 is a flowchart of a method for implementing an input prompt according to an embodiment of the present invention;
图2为本发明实施例另一种实现输入提示的方法(服务器)的流程图；2 is a flowchart of another method (server) for implementing an input prompt according to an embodiment of the present invention;
图3为本发明实施例另一种实现输入提示的方法(终端)的流程图；FIG. 3 is a flowchart of another method (terminal) for implementing an input prompt according to an embodiment of the present invention;
图4为本发明实施例终端的结构组成示意图；4 is a schematic structural diagram of a terminal according to an embodiment of the present invention;
图5为本发明实施例服务器的结构组成示意图；FIG. 5 is a schematic structural diagram of a server according to an embodiment of the present invention; FIG.
图6为本发明实施例另一种终端的结构组成示意图。FIG. 6 is a schematic structural diagram of another terminal according to an embodiment of the present invention.
具体实施方式detailed description
为了便于本领域技术人员的理解，下面结合附图对本发明作进一步的描述，并不能用来限制本发明的保护范围。需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的各种方式可以相互组合。In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings, and is not intended to limit the scope of the present invention. It should be noted that the embodiments in the present application and the various manners in the embodiments may be combined with each other without conflict.
参见图1，本发明实施例提出了一种实现输入提示的方法，包括：Referring to FIG. 1, an embodiment of the present invention provides a method for implementing an input prompt, including:
步骤100、终端采集语音信号，将采集的语音信号转换为文本文字。 Step 100: The terminal collects a voice signal, and converts the collected voice signal into text text.
本步骤中，终端可以采集语音模拟信号，将语音模拟信号转换成语音数字信号，对语音数字信号进行滤波，将滤波后的语音数字信号转换成文本文字，具体可以采用本领域技术人员的公知技术实现，并不用于限定本发明的保护范围，这里不再赘述。In this step, the terminal can collect the voice analog signal, convert the voice analog signal into a voice digital signal, filter the voice digital signal, and convert the filtered voice digital signal into text text, which can be specifically known by those skilled in the art. The implementation is not intended to limit the scope of the present invention, and details are not described herein.
步骤101、终端提取所述文本文字中的词。Step 101: The terminal extracts words in the text text.
本步骤中，所述终端提取所述文本文字中的词，包括：In this step, the terminal extracts words in the text text, including:
终端将文本文字以词为单位进行划分得到文本文字中的词。The terminal divides the text text into words in units of words to obtain words in the text.
其中，终端将文本文字以词为单位进行划分得到文本文字中的词，包括：终端确定划分方案，按照确定的划分方案将文本文字进行划分得到文本文字中的词。The terminal divides the text text into words in units of words to obtain words in the text text, including: the terminal determines the division scheme, and divides the text text according to the determined division scheme to obtain words in the text text.
其中，所述终端确定划分方案包括：The determining the division scheme by the terminal includes:
终端计算文本文字中的每一个字对应的状态向量，计算状态转移矩阵，根据所有字对应的状态向量和状态转移矩阵计算每一种分词方案的分数，确定分数最大的分词方案为划分方案。The terminal calculates a state vector corresponding to each word in the text text, calculates a state transition matrix, calculates a score of each word segmentation scheme according to a state vector corresponding to all words and a state transition matrix, and determines a word segmentation scheme with the largest score as a division scheme.
其中，状态向量中的每一个元素表示该字为该元素对应的状态的概率。Wherein each element in the state vector represents the probability that the word is the state corresponding to the element.
其中，终端计算文本文字中的每一个字对应的状态向量包括：Wherein, the terminal calculates a state vector corresponding to each word in the text text, including:
终端将文本文字中的每一个字转换成字向量，根据字向量计算该字的状态向量。The terminal converts each word in the text into a word vector, and calculates a state vector of the word based on the word vector.
其中，终端将文本文字中的每一个字转换成字向量包括：终端在预先设置的字和字向量之间的第一对应关系中查找文本文字中的每一个字对应的字向量。The terminal converting each word in the text text into a word vector includes: the terminal searching for a word vector corresponding to each word in the text text in a first correspondence between the preset word and the word vector.
其中，终端可以根据字向量采用多层神经网络模型计算该字的状态向量。The terminal can calculate the state vector of the word according to the word vector using a multi-layer neural network model.
其中，具体如何计算状态转移矩阵可以采用本领域技术人员的公知技术实现，并不用于限定本发明的保护范围，这里不再赘述。Specifically, how to calculate the state transition matrix may be implemented by using a well-known technology of the person skilled in the art, and is not intended to limit the scope of protection of the present invention, and details are not described herein again.
例如，一个字对应的状态可以是B，或E，或I，或S，其中，B表示 词的开头，E表示词的结尾，I表示词的中间，S表示一个词。For example, the state corresponding to a word can be B, or E, or I, or S, where B represents At the beginning of the word, E is the end of the word, I is the middle of the word, and S is a word.
则状态向量可以表示为(PB，PE，PI，PS)，其中，PB表示该字是词的开头的概率，PE表示该字是词的结尾的概率，PI表示该字是词的中间的概率，PS表示该字是一个词的概率。Then the state vector can be expressed as (P B , P E , P I , P S ), where P B represents the probability that the word is the beginning of the word, P E represents the probability that the word is the end of the word, P I represents the probability The word is the probability in the middle of the word, and P S is the probability that the word is a word.
如果文本文字为“他爱打球”，则计算得到的“他”的状态向量为(0.1,0.1,0.1,0.7)，“爱”的状态向量为(0.2,0.1,0,0.7)，“打”的状态向量为(0.6,0.2，0.1,0.1)，“球”的状态向量为(0.1,0.8,0.1,0)。If the text is "he loves to play", the calculated state vector of "he" is (0.1, 0.1, 0.1, 0.7), and the state vector of "love" is (0.2, 0.1, 0, 0.7), "play" The state vector of "" is (0.6, 0.2, 0.1, 0.1), and the state vector of "ball" is (0.1, 0.8, 0.1, 0).
状态转移矩阵为
Figure PCTCN2017070905-appb-000003
State transition matrix is
Figure PCTCN2017070905-appb-000003
其中，根据所有字对应的状态向量和状态转移矩阵计算每一种分词方案的分数包括：Wherein, the scores of each word segmentation scheme are calculated according to the state vector and the state transition matrix corresponding to all words:
按照公式
Figure PCTCN2017070905-appb-000004
计算第i种分词方案的分数；According to the formula
Figure PCTCN2017070905-appb-000004
Calculate the score of the i-th word segmentation scheme;
其中，si为第i种分词方案的分数，Aij为第i种分词方案第(j－1)个字的状态转移到第j个字的状态的概率，fij为第i种分词方案第j个字的状态。Where s i is the score of the i-th word segmentation scheme, A ij is the probability that the state of the (i-1)th word of the i-th participle program shifts to the state of the j-th word, and f ij is the i-th word-dividing scheme The state of the jth word.
步骤102、终端获取与文本文字中的词相关联的词并显示。Step 102: The terminal acquires a word associated with a word in the text text and displays the word.
作为一种实施方式，步骤101和步骤102之间还包括：As an implementation manner, between step 101 and step 102, the method further includes:
终端确定文本文字中的词的词性；相应地，The terminal determines the part of speech of the word in the text text; accordingly,
终端获取与文本文字中的词相近的词并显示包括：终端获取与文本文字中除虚词以外的词相近的词并显示。The terminal acquires a word similar to the word in the text text and displays that the terminal acquires and displays a word similar to the word other than the virtual word in the text text.
其中，终端确定文本文字中的词的词性包括：Wherein, the terminal determines the part of speech of the word in the text:
终端计算文本文字中的每一个词对应的状态向量，计算状态转移矩阵，根据所有词对应的状态向量和状态转移矩阵计算每一种确定词性的方案的分数，根据分数最大的确定词性的方案确定文本文字中的每一个词的词性。 The terminal calculates a state vector corresponding to each word in the text text, calculates a state transition matrix, calculates a score of each program for determining the part of speech according to the state vector and the state transition matrix corresponding to all words, and determines according to the scheme of determining the part of speech with the largest score. The part of speech of each word in the text.
上述的具体实现过程与终端将文本文字以词为单位进行划分得到文本文字中的词类似，具体可以参考上面的实现过程来实现。The specific implementation process described above and the terminal divide the text text into words in units of words to obtain similar words in the text text, which can be specifically implemented by referring to the above implementation process.
步骤102中，终端获取与文本文字中的词相关联的词并显示包括：In step 102, the terminal acquires a word associated with the word in the text text and displays:
终端获取与文本文字中的每一个词或除虚词以外的每一个词最相近的N个词；其中，N为大于或等于1的整数；终端计算获得的M×N个词的得分；M为文本文字中的词的个数或除虚词以外的词的个数；终端按照得分从高到低的顺序进行显示。The terminal acquires N words that are the closest to each word in the text text or each word except the virtual word; wherein N is an integer greater than or equal to 1; the terminal calculates the score of the obtained M×N words; The number of words in the text or the number of words other than the virtual word; the terminal displays in descending order of the score.
其中，终端获取与文本文字中的每一个词或除虚词以外的每一个词最相近的N个词包括：The N words that the terminal obtains the most similar to each word in the text or each word except the virtual word include:
终端将文本文字中的每一个词或除虚词以外的每一个词转换成第一词向量，获取与每一个第一词向量距离最小的N个第二词向量对应的词。The terminal converts each word in the text text or each word except the function word into a first word vector, and acquires a word corresponding to the N second word vectors having the smallest distance from each of the first word vectors.
其中，终端将文本文字中的每一个词或除虚词以外的每一个词转换成第一词向量包括：Wherein, the terminal converts each word in the text text or each word except the function word into the first word vector including:
终端在预先设置的词和词向量之间的第二对应关系中，查找文本文字中的每一个词或除虚词以外的每一个词对应的第一词向量。The terminal searches for each word in the text text or the first word vector corresponding to each word except the virtual word in the second correspondence between the preset word and the word vector.
其中，获取与每一个第一词向量距离最小的N个第二词向量对应的词包括：Wherein, the words corresponding to the N second word vectors having the smallest distance from each of the first word vectors are:
终端计算每一个第一词向量与预先设置的词向量空间中的每一个第二词向量之间的距离，获取将计算得到的距离最小的N个第二向量对应的词。The terminal calculates a distance between each of the first word vectors and each of the second word vectors in the preset word vector space, and acquires words corresponding to the N second vectors that minimize the calculated distance.
上述方法中，采用词向量来表示词，那么得到的与文本文字相近的词的第一个字不一定与用户输入的字相同，从而联想的词更加丰富。In the above method, the word vector is used to represent the word, and the first word of the word that is similar to the text is not necessarily the same as the word input by the user, so that the associated word is more abundant.
其中，假设词向量的起点为空间坐标原点，可以用两向量之间的夹角来表示两向量之间的距离，即终端计算每一个第一词向量与预先设置的词向量空间中的每一个第二词向量之间的距离包括：Wherein, assuming that the starting point of the word vector is the origin of the space coordinate, the distance between the two vectors can be represented by the angle between the two vectors, that is, the terminal calculates each of the first word vector and each of the preset word vector spaces. The distance between the second word vectors includes:
终端按照公式
Figure PCTCN2017070905-appb-000005
计算第一词向量和第二词向量之间的距离； Terminal according to formula
Figure PCTCN2017070905-appb-000005
Calculating a distance between the first word vector and the second word vector;
其中，θ为所述第一词向量和所述第二词向量之间的夹角，用于表示所述第一词向量和所述第二词向量之间的距离，a为第一词向量，b为所述第二词向量。θ的取值范围为0到π；Where θ is an angle between the first word vector and the second word vector for indicating a distance between the first word vector and the second word vector, a is a first word vector , b is the second word vector. θ ranges from 0 to π;
或者，所述终端按照公式d＝||a-b||p计算所述第一词向量和所述第二词向量之间的距离；p＝(0,+∞]；Alternatively, the terminal calculates a distance between the first word vector and the second word vector according to the formula d=||ab|| p ; p=(0, +∞);
其中，d为第一词向量和第二词向量之间的距离，a为所述第一词向量，b为所述第二词向量；当p＝1时，d表示第一词向量和所述第二词向量之间的曼哈顿距离，当p＝2时，d表示第一词向量和所述第二词向量之间的欧式距离。当p→∞时，d表示第一词向量和所述第二词向量之间的切比雪夫距离。Where d is the distance between the first word vector and the second word vector, a is the first word vector, b is the second word vector; when p=1, d represents the first word vector and The Manhattan distance between the second word vectors, when p=2, d represents the Euclidean distance between the first word vector and the second word vector. When p→∞, d represents the Chebyshev distance between the first word vector and the second word vector.
其中，终端计算获得的M×N个词的得分包括：The scores of the M×N words obtained by the terminal calculation include:
终端计算M×N个词中每一个词的分数，将M×N个词中出现多次的词对应的分数相加得到的和值作为该词的得分；将M×N个词中出现一次的词对应的分数作为该词的得分。The terminal calculates the score of each of the M×N words, and adds the sum of the scores corresponding to the words that appear multiple times in the M×N words as the score of the word; appears once in the M×N words The word corresponds to the score as the score of the word.
其中，终端计算M×N个词中每一个词的分数包括：Wherein, the terminal calculates the score of each of the M×N words including:
终端按照公式c＝wL计算所述M×N个词中每一个词的分数；The terminal calculates a score of each of the M×N words according to the formula c=wL;
其中，c为M×N个词中每一个词的分数，w为M×N个词中每一个词对应的第一词向量对应的词的权重，L为第一词向量和M×N个词中每一个词对应的所述第二词向量之间的距离。Where c is the score of each of the M×N words, w is the weight of the word corresponding to the first word vector corresponding to each of the M×N words, and L is the first word vector and M×N The distance between the second word vectors corresponding to each word in the word.
其中，第一词向量对应的词的权值为
Figure PCTCN2017070905-appb-000006
其中，s为第一词向量对应的词在文本文字中的位置。第一词向量对应的词在文本文字中的位置是指，将文本文字中的所有词或除虚词以外的词从头到尾进行编号后，第一词向量对应的词的编号。例如，文本文字为“他爱打球”，分成三个词：他、爱、打球，将“他”编号为1，“爱”编号为2，“打球”编号为3；或者将“爱”编号为1，“打球”编号为2。 Wherein the weight of the word corresponding to the first word vector is
Figure PCTCN2017070905-appb-000006
Where s is the position of the word corresponding to the first word vector in the text text. The position of the word corresponding to the first word vector in the text text refers to the number of the word corresponding to the first word vector after all words in the text text or words other than the virtual word are numbered from beginning to end. For example, the text is "He loves to play" and is divided into three words: he, love, play, number "he" as 1, "love" number 2, "play" number 3; or "love" number For 1, the "playing" number is 2.
参见图2，本发明实施例还提出了一种实现输入提示的方法，包括：Referring to FIG. 2, an embodiment of the present invention further provides a method for implementing an input prompt, including:
步骤200、服务器接收到来自终端采集的语音信号，将接收到的语音信号转换为文本文字。Step 200: The server receives the voice signal collected by the terminal, and converts the received voice signal into text text.
本步骤中，服务器如何将接收到的语音信号转换为文本文字可以采用本领域技术人员的公知技术实现，并不用于限定本发明的保护范围，这里不再赘述。In this step, how the server converts the received voice signal into text text can be implemented by using techniques well known to those skilled in the art, and is not intended to limit the scope of protection of the present invention, and details are not described herein again.
步骤201、服务器获取文本文字中的词。Step 201: The server acquires a word in the text text.
本步骤的具体实现方式与步骤101相同。The specific implementation of this step is the same as step 101.
步骤202、服务器获取与所述文本文字中的词相关联的词并发送给终端。Step 202: The server acquires a word associated with the word in the text text and sends the word to the terminal.
本步骤中，服务器获取与文本文字中的词相近的词与步骤102中终端获取与文本文字中的词相近的词的实现方式相同。In this step, the server acquires a word similar to the word in the text text and the terminal in step 102 acquires the word similar to the word in the text text.
本步骤中，服务器还可以将与文本文字中的词相近的词的得分发送给终端。或者，服务器按照与文本文字中的词性相近的词的得分从高到低的顺序进行排序，将排序的与文本文字中的词相近的词发送给终端。In this step, the server may also send a score of a word similar to the word in the text text to the terminal. Alternatively, the server sorts the scores of the words that are similar to the part of speech in the text text from high to low, and sends the sorted words that are similar to the words in the text text to the terminal.
作为一种实施方式，步骤201和步骤202之间还包括：As an implementation manner, between step 201 and step 202, the method further includes:
服务器确定文本文字中的词的词性；相应地，The server determines the part of speech of the word in the text text; accordingly,
服务器获取与文本文字中的词相近的词并发送给终端包括：The server obtains words similar to the words in the text and sends them to the terminal including:
服务器获取与文本文字中除虚词以外的词相近的词并发送给所述终端。The server acquires a word similar to a word other than the virtual word in the text text and transmits the word to the terminal.
其中，服务器确定文本文字中的词的词性与终端确定文本文字中的词的词性的实现方式相同。Wherein, the server determines that the part of speech of the word in the text text is implemented in the same manner as the word form of the word in the terminal determining the text word.
参见图3，本发明实施例还提出了一种实现输入提示的方法，包括：Referring to FIG. 3, an embodiment of the present invention further provides a method for implementing an input prompt, including:
步骤300、终端采集语音信号，将采集的语音信号发送给服务器。Step 300: The terminal collects a voice signal, and sends the collected voice signal to the server.
步骤301、终端接收到所述服务器发送的与所述语音信号对应的文本文字中的词相关联的词并显示。Step 301: The terminal receives and displays a word associated with a word in a text text corresponding to the voice signal sent by the server.
本步骤中，终端还可以按照接与文本文字中的词相近的词的得分从高到底的顺序进行显示。 In this step, the terminal can also display the scores of words that are close to the words in the text text from high to low.
参见图4，本发明实施例还提出了一种终端，包括：Referring to FIG. 4, an embodiment of the present invention further provides a terminal, including:
第一采集模块，配置为采集语音信号；The first collection module is configured to collect a voice signal;
第一转换模块，配置为将采集的语音信号转换为文本文字；a first conversion module configured to convert the collected voice signal into text text;
第一处理模块，配置为获取文本文字中的词；获取与文本文字中的词相近的词并显示。The first processing module is configured to acquire words in the text text; acquire words that are similar to the words in the text text and display them.
本发明实施例的终端中，所述第一处理模块配置为：In the terminal of the embodiment of the present invention, the first processing module is configured to:
获取文本文字中的词；确定所述文本文字中的词的词性；获取与所述文本文字中除虚词以外的词相近的词并显示。Obtaining words in the text text; determining the part of speech of the words in the text text; acquiring words that are similar to words other than the virtual words in the text text and displaying them.
本发明实施例中，所述终端中的第一采集模块、第一转换模块和第一处理模块，在实际应用中均可由所述终端中的中央处理器(CPU，Central Processing Unit)、数字信号处理器(DSP，Digital Signal Processor)、微控制单元(MCU，Microcontroller Unit)或可编程门阵列(FPGA，Field－Programmable Gate Array)实现。In the embodiment of the present invention, the first acquisition module, the first conversion module, and the first processing module in the terminal may be used by a central processing unit (CPU) and a digital signal in the terminal in an actual application. A DSP (Digital Signal Processor), a Microcontroller Unit (MCU), or a Field-Programmable Gate Array (FPGA).
参见图5，本发明实施例还提出了一种服务器，包括：Referring to FIG. 5, an embodiment of the present invention further provides a server, including:
第一接收模块，配置为收到来自终端采集的语音信号；a first receiving module configured to receive a voice signal collected from the terminal;
第二转换模块，配置为将接收到的语音信号转换为文本文字；a second conversion module configured to convert the received voice signal into text text;
第二处理模块，配置为获取文本文字中的词；获取与文本文字中的词相近的词；a second processing module configured to obtain a word in the text text; to obtain a word that is similar to the word in the text text;
第一发送模块，将与文本文字中的词相近的词发送给终端。The first sending module sends a word similar to the word in the text text to the terminal.
本发明实施例的服务器中，所述第二处理模块配置为：In the server of the embodiment of the present invention, the second processing module is configured to:
获取文本文字中的词；确定所述文本文字中的词的词性；获取与所述文本文字中除虚词以外的词相近的词。Obtaining words in the text text; determining the part of speech of the words in the text text; and acquiring words similar to words other than the virtual words in the text text.
本发明实施例中，所述服务器中的第二转换模块和第二处理模块，在实际应用中均可由所述服务器中的CPU、DSP、MCU或FPGA实现；所述服务器中的第一接收模块和第一发送模块，在实际应用中可通过通信模组(包含：基础通信套件、操作系统、通信模块、标准化接口和协议等)及 收发天线实现。In the embodiment of the present invention, the second conversion module and the second processing module in the server may be implemented by a CPU, a DSP, an MCU or an FPGA in the server in an actual application; a first receiving module in the server And the first sending module, in practical applications, can pass the communication module (including: basic communication suite, operating system, communication module, standardized interface and protocol, etc.) Transceiver antenna implementation.
参见图6，本发明实施例还提出了一种终端，包括：Referring to FIG. 6, an embodiment of the present invention further provides a terminal, including:
第二采集模块，配置为采集语音信号；a second collection module configured to collect a voice signal;
第二发送模块，配置为将采集的语音信号发送给服务器；a second sending module, configured to send the collected voice signal to the server;
第二接收模块，配置为接收到来自服务器的与语音信号对应的文本文字中的词相近的词并显示。The second receiving module is configured to receive and display a word similar to the word in the text text corresponding to the voice signal from the server.
本发明实施例中，所述终端中的第二采集模块，在实际应用中可由所述终端中的CPU、DSP、MCU或FPGA实现；所述终端中的第二发送模块和第二接收模块，在实际应用中可通过通信模组(包含：基础通信套件、操作系统、通信模块、标准化接口和协议等)及收发天线实现。In the embodiment of the present invention, the second collection module in the terminal may be implemented by a CPU, a DSP, an MCU, or an FPGA in the terminal in a practical application; a second sending module and a second receiving module in the terminal, In practical applications, it can be realized through communication modules (including: basic communication suite, operating system, communication module, standardized interface and protocol, etc.) and transceiver antennas.
在本申请所提供的几个实施例中，应该理解到，所揭露的设备和方法，可以通过其它的方式实现。以上所描述的设备实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，如：多个单元或组件可以结合，或可以集成到另一个系统，或一些特征可以忽略，或不执行。另外，所显示或讨论的各组成部分相互之间的耦合、或直接耦合、或通信连接可以是通过一些接口，设备或单元的间接耦合或通信连接，可以是电性的、机械的或其它形式的。In the several embodiments provided by the present application, it should be understood that the disclosed apparatus and method may be implemented in other manners. The device embodiments described above are merely illustrative. For example, the division of the unit is only a logical function division. In actual implementation, there may be another division manner, such as: multiple units or components may be combined, or Can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, or direct coupling, or communication connection of the components shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, and may be electrical, mechanical or other forms. of.
上述作为分离部件说明的单元可以是、或也可以不是物理上分开的，作为单元显示的部件可以是、或也可以不是物理单元，即可以位于一个地方，也可以分布到多个网络单元上；可以根据实际的需要选择其中的部分或全部单元来实现本实施例方案的目的。The units described above as separate components may or may not be physically separated, and the components displayed as the unit may or may not be physical units, that is, may be located in one place or distributed to multiple network units; Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
另外，在本发明各实施例中的各功能单元可以全部集成在一个处理单元中，也可以是各单元分别单独作为一个单元，也可以两个或两个以上单元集成在一个单元中；上述集成的单元既可以采用硬件的形式实现，也可以采用硬件加软件功能单元的形式实现。In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated into one unit; The unit can be implemented in the form of hardware or in the form of hardware plus software functional units.
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步 骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：移动存储设备、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。One of ordinary skill in the art can understand that all or part of the steps of the above method embodiments are implemented. The foregoing may be completed by a program instruction related hardware, where the foregoing program may be stored in a computer readable storage medium, and when executed, the program includes the steps of the foregoing method embodiment; and the foregoing storage medium includes: mobile storage A device that can store program code, such as a device, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.
或者，本发明上述集成的单元如果以软件功能模块的形式实现并作为独立的产品销售或使用时，也可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实施例的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机、服务器、或者网络设备等)执行本发明各个实施例所述方法的全部或部分。而前述的存储介质包括：移动存储设备、ROM、RAM、磁碟或者光盘等各种可以存储程序代码的介质。Alternatively, the above-described integrated unit of the present invention may be stored in a computer readable storage medium if it is implemented in the form of a software function module and sold or used as a standalone product. Based on such understanding, the technical solution of the embodiments of the present invention may be embodied in the form of a software product in essence or in the form of a software product stored in a storage medium, including a plurality of instructions. A computer device (which may be a personal computer, server, or network device, etc.) is caused to perform all or part of the methods described in various embodiments of the present invention. The foregoing storage medium includes various media that can store program codes, such as a mobile storage device, a ROM, a RAM, a magnetic disk, or an optical disk.
以上所述，仅为本发明的具体实施方式，但本发明的保护范围并不局限于此，任何熟悉本技术领域的技术人员在本发明揭露的技术范围内，可轻易想到变化或替换，都应涵盖在本发明的保护范围之内。因此，本发明的保护范围应以所述权利要求的保护范围为准。The above is only a specific embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the present invention. It should be covered by the scope of the present invention. Therefore, the scope of the invention should be determined by the scope of the appended claims.
工业实用性Industrial applicability
本发明实施例的技术方案通过终端采集语音信号，将采集的语音信号转换为文本文字；提取所述文本文字中的词；获取与所述文本文字中的词相关联的词并显示，实现了对输入的词语的联想。 The technical solution of the embodiment of the present invention collects a voice signal by using a terminal, converts the collected voice signal into text text, extracts a word in the text text, acquires a word associated with the word in the text text, and displays the The association of the words entered.

Claims (24)

  1. 一种实现输入提示的方法，包括：A method of implementing input prompts, including:
    终端采集语音信号，将采集的语音信号转换为文本文字；The terminal collects a voice signal, and converts the collected voice signal into text text;
    终端提取所述文本文字中的词；The terminal extracts words in the text text;
    终端获取与所述文本文字中的词相关联的词并显示。The terminal acquires and displays the words associated with the words in the text text.
  2. 根据权利要求1所述的方法，其中，所述终端提取所述文本文字中的词后，在所述终端获取与所述文本文字中的词相关联的词并显示之前还包括：The method according to claim 1, wherein the terminal extracts a word in the text text, and before the terminal acquires a word associated with the word in the text text and displays the word, the method further includes:
    所述终端确定所述文本文字中的词的词性；The terminal determines the part of speech of the word in the text text;
    所述终端获取与文本文字中的词相近的词并显示包括：The terminal acquires a word similar to the word in the text text and displays:
    所述终端获取与所述文本文字中除虚词以外的词相关联的词并显示。The terminal acquires and displays a word associated with a word other than the virtual word in the text text.
  3. 根据权利要求1或2所述的方法，其中，所述终端提取所述文本文字中的词，包括：The method according to claim 1 or 2, wherein the terminal extracts words in the text text, including:
    所述终端将所述文本文字以词为单位进行划分得到所述文本文字中的词。The terminal divides the text text into words in units of words to obtain words in the text text.
  4. 根据权利要求3所述的方法，其中，所述终端将文本文字以词为单位进行划分得到文本文字中的词，包括：The method according to claim 3, wherein the terminal divides the text text into words in units of words to obtain words in the text, including:
    所述终端确定划分方案，按照确定的划分方案将所述文本文字进行划分得到所述文本文字中的词。The terminal determines a division scheme, and divides the text text according to the determined division scheme to obtain words in the text text.
  5. 根据权利要求4所述的方法，其中，所述终端确定划分方案包括：The method of claim 4, wherein the determining the partitioning scheme by the terminal comprises:
    所述终端计算文本文字中的每一个字对应的状态向量，计算状态转移矩阵，根据所有字对应的状态向量和状态转移矩阵计算每一种分词方案的分数，确定分数最大的分词方案为所述划分方案。The terminal calculates a state vector corresponding to each word in the text text, calculates a state transition matrix, calculates a score of each word segmentation scheme according to a state vector corresponding to all words and a state transition matrix, and determines a word segmentation scheme with the largest score as described Division plan.
  6. 根据权利要求1或2所述的方法，其中，所述终端获取与所述文本文字中的词相关联的词并显示包括： The method according to claim 1 or 2, wherein the terminal acquires a word associated with the word in the text text and displays:
    所述终端获取与所述文本文字中的每一个词或除虚词以外的每一个词最相近的N个词；其中，N为大于或等于1的整数；The terminal acquires N words that are the closest to each word in the text text or each word except the virtual word; wherein N is an integer greater than or equal to 1;
    所述终端计算获得的M×N个词的得分；M为所述文本文字中的词的个数或除虚词以外的词的个数；The terminal calculates a score of the obtained M×N words; M is the number of words in the text text or the number of words other than the virtual word;
    所述终端按照得分从高到低的顺序进行显示。The terminals are displayed in descending order of score.
  7. 根据权利要求6所述的方法，其中，所述终端获取与文本文字中的每一个词或除虚词以外的每一个词最相近的N个词包括：The method of claim 6, wherein the terminal acquires N words that are most similar to each word in the text or to each word except the word:
    所述终端将所述文本文字中的每一个词或除虚词以外的每一个词转换成第一词向量，获取与每一个第一词向量距离最小的N个第二词向量对应的词。The terminal converts each word in the text text or each word except the virtual word into a first word vector, and acquires a word corresponding to the N second word vectors having the smallest distance from each of the first word vectors.
  8. 根据权利要求7所述的方法，其中，所述终端将文本文字中的每一个词或除虚词以外的每一个词转换成第一词向量包括：The method according to claim 7, wherein the terminal converting each word in the text text or each word other than the function word into the first word vector comprises:
    所述终端在预先设置的词和词向量之间的第二对应关系中，查找所述文本文字中的每一个词或所述除虚词以外的每一个词对应的第一词向量。The terminal searches for each word in the text character or the first word vector corresponding to each word except the virtual word in a second correspondence between the preset word and the word vector.
  9. 根据权利要求7所述的方法，其中，所述获取与每一个第一词向量距离最小的N个第二词向量对应的词包括：The method of claim 7, wherein the obtaining words corresponding to the N second word vectors having the smallest distance from each of the first word vectors comprises:
    所述终端计算每一个所述第一词向量与预先设置的词向量空间中的每一个第二词向量之间的距离，获取将计算得到的距离最小的N个第二向量对应的词。The terminal calculates a distance between each of the first word vectors and each of the second word vectors in the preset word vector space, and acquires words corresponding to the N second vectors that minimize the calculated distance.
  10. 根据权利要求9所述的方法，其中，所述终端计算每一个第一词向量与预先设置的词向量空间中的每一个第二词向量之间的距离包括：The method of claim 9, wherein the terminal calculates a distance between each of the first word vectors and each of the second word vectors in the preset word vector space comprises:
    所述终端按照公式
    Figure PCTCN2017070905-appb-100001
    计算所述第一词向量和所述第二词向量之间的距离；    The terminal according to the formula
    Figure PCTCN2017070905-appb-100001
    Calculating a distance between the first word vector and the second word vector;
    其中，θ为所述第一词向量和所述第二词向量之间的夹角，用于表示所述第一词向量和所述第二词向量之间的距离，a为所述第一词向量，b为 所述第二词向量；Where θ is an angle between the first word vector and the second word vector, and is used to represent a distance between the first word vector and the second word vector, where a is the first Word vector, b is The second word vector;
    或者，所述终端按照公式d＝||a-b||p计算所述第一词向量和所述第二词向量之间的距离；p＝(0,+∞]；Alternatively, the terminal calculates a distance between the first word vector and the second word vector according to the formula d=||ab|| p ; p=(0, +∞);
    其中，d为所述第一词向量与所述第二词向量之间的距离，a为所述第一词向量，b为所述第二词向量；当p＝1时，d表示第一词向量和所述第二词向量之间的曼哈顿距离，当p＝2时，d表示第一词向量和所述第二词向量之间的欧式距离。当p→∞时，d表示第一词向量和所述第二词向量之间的切比雪夫距离。Where d is the distance between the first word vector and the second word vector, a is the first word vector, b is the second word vector; when p=1, d represents the first The Manhattan distance between the word vector and the second word vector, when p=2, d represents the Euclidean distance between the first word vector and the second word vector. When p→∞, d represents the Chebyshev distance between the first word vector and the second word vector.
  11. 根据权利要求9所述的方法，其中，所述终端计算获得的M×N个词的得分包括：The method according to claim 9, wherein the terminal calculates the scores of the obtained M x N words including:
    所述终端计算所述M×N个词中每一个词的分数，将所述M×N个词中出现多次的词对应的分数相加得到的和值作为该词的得分；将所述M×N个词中出现一次的词对应的分数作为该词的得分。The terminal calculates a score of each of the M×N words, and adds a sum value obtained by adding the scores corresponding to the plurality of words in the M×N words as a score of the word; The score corresponding to the word that appears once in M×N words is taken as the score of the word.
  12. 根据权利要求11所述的方法，其中，所述终端计算M×N个词中每一个词的分数包括：The method of claim 11, wherein the terminal calculates a score for each of the M x N words comprising:
    所述终端按照公式c＝wL计算所述M×N个词中每一个词的分数；The terminal calculates a score of each of the M×N words according to a formula c=wL;
    其中，c为所述M×N个词中每一个词的分数，w为M×N个词中每一个词对应的所述第一词向量对应的词的权重，L为所述第一词向量和M×N个词中每一个词对应的所述第二词向量之间的距离。Where c is the score of each of the M×N words, w is the weight of the word corresponding to the first word vector corresponding to each of the M×N words, and L is the first word The distance between the vector and the second word vector corresponding to each of the M x N words.
  13. 根据权利要求12所述的方法，其中，所述第一词向量对应的词的权值为
    Figure PCTCN2017070905-appb-100002
    其中，s为所述第一词向量对应的词在所述文本文字中的位置；所述S表示文本文字中所有词或除虚词以外的词进行编号后，所述第一词向量对应的词的编号。    The method of claim 12, wherein the weight of the word corresponding to the first word vector is
    Figure PCTCN2017070905-appb-100002
    Where s is the position of the word corresponding to the first word vector in the text text; and S represents a word corresponding to the first word vector after all words in the text character or words other than the virtual word are numbered The number.
  14. 一种实现输入提示的方法，包括：A method of implementing input prompts, including:
    服务器接收到来自终端采集的语音信号，将接收到的语音信号转换为 文本文字；The server receives the voice signal collected from the terminal, and converts the received voice signal into Text text
    服务器获取所述文本文字中的词；The server obtains a word in the text text;
    服务器获取与所述文本文字中的词相关联的词并发送给终端。The server obtains the words associated with the words in the text text and sends them to the terminal.
  15. 根据权利要求14所述的方法，其中，所述服务器获取所述文本文字中的词后，在所述服务器获取与所述文本文字中的词相关联的词并发送给终端之前还包括：The method according to claim 14, wherein after the server acquires the words in the text text, before the server acquires the words associated with the words in the text text and sends the words to the terminal, the method further includes:
    所述服务器确定所述文本文字中的词的词性；The server determines the part of speech of the word in the text text;
    所述服务器获取与文本文字中的词相近的词并发送给终端包括：The server acquires a word similar to the word in the text text and sends the word to the terminal, including:
    所述服务器获取与所述文本文字中除虚词以外的词相关联的词并发送给所述终端。The server acquires a word associated with a word other than the virtual word in the text text and transmits the word to the terminal.
  16. 一种实现输入提示的方法，包括：A method of implementing input prompts, including:
    终端采集语音信号，将采集的语音信号发送给服务器；The terminal collects a voice signal, and sends the collected voice signal to the server;
    终端接收到所述服务器发送的与所述语音信号对应的文本文字中的词相关联的词并显示。The terminal receives and displays a word associated with the word in the text text corresponding to the voice signal sent by the server.
  17. 一种终端，包括：A terminal comprising:
    第一采集模块，配置为采集语音信号；The first collection module is configured to collect a voice signal;
    第一转换模块，配置为将采集的语音信号转换为文本文字；a first conversion module configured to convert the collected voice signal into text text;
    第一处理模块，配置为提取所述文本文字中的词；获取与所述文本文字中的词相关联的词并显示。a first processing module configured to extract words in the textual text; to acquire and display words associated with words in the textual text.
  18. 根据权利要求17所述的终端，其中，所述第一处理模块配置为：The terminal of claim 17, wherein the first processing module is configured to:
    获取文本文字中的词；确定所述文本文字中的词的词性；获取与所述文本文字中除虚词以外的词相关联的词并显示。Obtaining words in the text text; determining the part of speech of the words in the text text; acquiring words associated with words other than the virtual words in the text text and displaying them.
  19. 一种服务器，包括：A server that includes:
    第一接收模块，配置为收到来自终端采集的语音信号；a first receiving module configured to receive a voice signal collected from the terminal;
    第二转换模块，配置为将接收到的语音信号转换为文本文字；a second conversion module configured to convert the received voice signal into text text;
    第二处理模块，配置为获取所述文本文字中的词；获取与所述文本文 字中的词相关联的词；a second processing module, configured to acquire a word in the text text; acquire and text the text a word associated with a word in a word;
    第一发送模块，配置为将与所述文本文字中的词相关联的词发送给终端。The first sending module is configured to send a word associated with the word in the text text to the terminal.
  20. 根据权利要求19所述的服务器，其中，所述第二处理模块配置为：The server of claim 19, wherein the second processing module is configured to:
    获取文本文字中的词；确定所述文本文字中的词的词性；获取与所述文本文字中除虚词以外的词相近的词。Obtaining words in the text text; determining the part of speech of the words in the text text; and acquiring words similar to words other than the virtual words in the text text.
  21. 一种终端，包括：A terminal comprising:
    第二采集模块，配置为采集语音信号；a second collection module configured to collect a voice signal;
    第二发送模块，配置为将采集的语音信号发送给服务器；a second sending module, configured to send the collected voice signal to the server;
    第二接收模块，配置为接收到所述服务器发送的与所述语音信号对应的文本文字中的词相关联的词并显示。The second receiving module is configured to receive and display a word associated with the word in the text text corresponding to the voice signal sent by the server.
  22. 一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行权利要求1至13任一项所述的实现输入提示的方法。A computer storage medium having stored therein computer executable instructions for performing the method of implementing an input prompt according to any one of claims 1 to 13.
  23. 一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行权利要求14或15所述的实现输入提示的方法。A computer storage medium having stored therein computer executable instructions for performing the method of implementing an input prompt according to claim 14 or 15.
  24. 一种计算机存储介质，所述计算机存储介质中存储有计算机可执行指令，所述计算机可执行指令用于执行权利要求16所述的实现输入提示的方法。 A computer storage medium having stored therein computer executable instructions for performing the method of implementing an input prompt of claim 16.
PCT/CN2017/070905 2016-03-14 2017-01-11 Input method, device, and computer storage medium WO2017157093A1 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201610144077.5A CN107193389A (en) 2016-03-14 2016-03-14 A kind of method and apparatus for realizing input
CN201610144077.5 2016-03-14

Publications (1)

Publication Number Publication Date
WO2017157093A1 true WO2017157093A1 (en) 2017-09-21

Family

ID=59852035

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2017/070905 WO2017157093A1 (en) 2016-03-14 2017-01-11 Input method, device, and computer storage medium

Country Status (2)

Country Link
CN (1) CN107193389A (en)
WO (1) WO2017157093A1 (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109918566A (en) * 2019-02-15 2019-06-21 Oppo广东移动通信有限公司 Querying method, device, electronic equipment and medium

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101655837A (en) * 2009-09-08 2010-02-24 北京邮电大学 Method for detecting and correcting error on text after voice recognition
CN102915733A (en) * 2011-11-17 2013-02-06 微软公司 Interactive speech recognition
CN104166462A (en) * 2013-05-17 2014-11-26 北京搜狗科技发展有限公司 Input method and system for characters
CN105336326A (en) * 2011-09-28 2016-02-17 苹果公司 Speech recognition repair using contextual information

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101770463A (en) * 2008-12-30 2010-07-07 英业达股份有限公司 Electronic device and method for querying and translating by utilizing voice input
KR101363866B1 (en) * 2013-03-13 2014-02-20 에스케이플래닛 주식회사 Method for generating of voice message, apparatus and system for the same

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101655837A (en) * 2009-09-08 2010-02-24 北京邮电大学 Method for detecting and correcting error on text after voice recognition
CN105336326A (en) * 2011-09-28 2016-02-17 苹果公司 Speech recognition repair using contextual information
CN102915733A (en) * 2011-11-17 2013-02-06 微软公司 Interactive speech recognition
CN104166462A (en) * 2013-05-17 2014-11-26 北京搜狗科技发展有限公司 Input method and system for characters

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109918566A (en) * 2019-02-15 2019-06-21 Oppo广东移动通信有限公司 Querying method, device, electronic equipment and medium
CN109918566B (en) * 2019-02-15 2023-05-23 Oppo广东移动通信有限公司 Query method, query device, electronic equipment and medium

Also Published As

Publication number Publication date
CN107193389A (en) 2017-09-22

Similar Documents

Publication Publication Date Title
CN107944559B (en) Method and system for automatically identifying entity relationship
CN107301865B (en) Method and device for determining interactive text in voice input
WO2018196718A1 (en) Image disambiguation method and device, storage medium, and electronic device
CN103019407B (en) Input method application method, automatic question answering processing method, electronic equipment and server
CN116797684B (en) Image generation method, device, electronic equipment and storage medium
WO2021051514A1 (en) Speech identification method and apparatus, computer device and non-volatile storage medium
CN112766229B (en) Human face point cloud image intelligent identification system and method based on attention mechanism
WO2023065731A1 (en) Method for training target map model, positioning method, and related apparatuses
WO2017005207A1 (en) Input method, input device, server and input system
US20220391587A1 (en) Method of training image-text retrieval model, method of multimodal image retrieval, electronic device and medium
CN113407850B (en) Method and device for determining and acquiring virtual image and electronic equipment
WO2017157093A1 (en) Input method, device, and computer storage medium
CN111582039B (en) Sign language recognition and conversion system and method based on deep learning and big data
CN109829040A (en) A kind of Intelligent dialogue method and device
US20240013558A1 (en) Cross-modal feature extraction, retrieval, and model training method and apparatus, and medium
CN115906861B (en) Sentence emotion analysis method and device based on interaction aspect information fusion
CN110895555B (en) Data retrieval method and device, storage medium and electronic device
EP4075424B1 (en) Speech recognition method and apparatus
WO2023000972A1 (en) Structured information extraction method and apparatus, and device and storage medium
CN109857886A (en) A kind of method for searching three-dimension model approached based on minimax value theory of games view
CN117011929A (en) Head posture estimation method, device, equipment and storage medium
CN114510942A (en) Method for acquiring entity words, and method, device and equipment for training model
CN113157852A (en) Voice processing method, system, electronic equipment and storage medium
CN113378774A (en) Gesture recognition method, device, equipment, storage medium and program product
CN112801078A (en) Point of interest (POI) matching method and device, electronic equipment and storage medium

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 17765635

Country of ref document: EP

Kind code of ref document: A1

122 Ep: pct application non-entry in european phase

Ref document number: 17765635

Country of ref document: EP

Kind code of ref document: A1