CN111462754A - Method for establishing dispatching control voice recognition model of power system - Google Patents
Method for establishing dispatching control voice recognition model of power system Download PDFInfo
- Publication number
- CN111462754A CN111462754A CN202010299229.5A CN202010299229A CN111462754A CN 111462754 A CN111462754 A CN 111462754A CN 202010299229 A CN202010299229 A CN 202010299229A CN 111462754 A CN111462754 A CN 111462754A
- Authority
- CN
- China
- Prior art keywords
- text data
- power
- dispatching
- power dispatching
- control instruction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000012795 verification Methods 0.000 claims abstract description 42
- 230000005236 sound signal Effects 0.000 claims abstract description 28
- 238000000605 extraction Methods 0.000 description 6
- 238000006467 substitution reaction Methods 0.000 description 2
- 241001672694 Citrus reticulata Species 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003058 natural language processing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- 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
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
-
- 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
- G10L17/00—Speaker identification or verification techniques
- G10L17/02—Preprocessing operations, e.g. segment selection; Pattern representation or modelling, e.g. based on linear discriminant analysis [LDA] or principal components; Feature selection or extraction
-
- 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
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/223—Execution procedure of a spoken command
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Multimedia (AREA)
- Acoustics & Sound (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Primary Health Care (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Computational Linguistics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a method for establishing a dispatching control voice recognition model of an electric power system, which comprises the steps of obtaining audio signals of personnel with electric power dispatching authority, obtaining corresponding voiceprints and establishing an electric power dispatching identity verification database; the method comprises the steps of obtaining voice data corresponding to each power dispatching control instruction, processing the voice data to obtain characteristic data and text data, comparing the text data with a preset power dispatching control instruction special dictionary to obtain target text data, establishing a corresponding relation between each characteristic data and the corresponding target text data, and forming a power system dispatching control voice recognition model through a power dispatching identity verification database and the corresponding relation. The power system dispatching control voice recognition model is used for power system dispatching, the voice recognition accuracy and reliability in the power system dispatching field can be improved according to the voice recognition model, and the requirement of power system dispatching with higher and higher intelligent requirements can be met.
Description
Technical Field
The invention relates to a method for establishing a scheduling control voice recognition model of an electric power system.
Background
Speech recognition is a technique for converting speech into text, and is a branch of artificial intelligence natural language processing. At present, the power system scheduling mainly adopts a manual scheduling mode, and the requirement of power system scheduling with higher and higher intelligent requirements is difficult to meet.
Disclosure of Invention
The invention aims to provide a method for establishing a speech recognition model for dispatching control of an electric power system, which is used for solving the problem that the existing electric power system dispatching mainly adopts a manual dispatching mode and is difficult to meet the dispatching requirement of the electric power system with higher and higher intelligent requirements.
In order to solve the problems, the invention adopts the following technical scheme:
a power system dispatching control voice recognition model building method comprises the following steps:
acquiring audio signals of personnel with power dispatching authority;
acquiring voiceprints of people corresponding to the audio signals according to the audio signals;
establishing a power dispatching identity verification database according to voiceprints of personnel with power dispatching authority;
acquiring voice data corresponding to each power dispatching control instruction;
extracting the characteristics of the voice data to obtain corresponding characteristic data;
recognizing the voice data to obtain corresponding text data;
comparing the text data with a preset special power dispatching control instruction dictionary, and if the text data belongs to words and sentences in the special power dispatching control instruction dictionary, retaining the text data to obtain target text data; if the text data does not belong to words and sentences in the electric power dispatching control instruction special dictionary, searching words and sentences in the electric power dispatching control instruction special dictionary which are the same as the pinyin of the text data according to the pinyin of the text data, and replacing the text data with the searched words and sentences to obtain target text data;
establishing a corresponding relation between each characteristic data and corresponding target text data;
and the power dispatching identity verification database and the corresponding relation form a power system dispatching control voice recognition model.
Optionally, after obtaining the voiceprint of the person corresponding to the audio signal, the method for establishing the power system scheduling control speech recognition model further includes:
acquiring face images of personnel with power dispatching authority, and establishing corresponding relations between different voiceprints and corresponding face images;
correspondingly, the establishing of the power dispatching identity verification database according to the voiceprints of the personnel with the power dispatching authority comprises the following steps:
and establishing a power dispatching identity verification database according to the voiceprint and the corresponding face image of each person with the power dispatching authority, wherein the power dispatching identity verification database comprises the voiceprint and the corresponding face image of each person with the power dispatching authority.
Optionally, the power dispatching control instruction special dictionary includes words and sentences of each power dispatching control instruction and pinyin of the words and sentences of each power dispatching control instruction;
the searching for the word and sentence in the power scheduling control instruction special dictionary which is the same as the pinyin of the text data according to the pinyin of the text data, and replacing the searched word and sentence with the text data to obtain target text data comprises:
obtaining pinyin of the text data;
inputting the obtained pinyin of the text data into the power dispatching control instruction special dictionary to obtain the pinyin which is the same as the pinyin of the text data in the power dispatching control instruction special dictionary, and obtaining corresponding words and sentences in the power dispatching control instruction special dictionary according to the obtained pinyin;
and replacing the text data with the obtained words and sentences to obtain target text data.
The invention has the beneficial effects that: the established power system dispatching control voice recognition model comprises a power dispatching identity verification database, the power dispatching identity verification database comprises voiceprints of personnel with power dispatching authority, then before voice corresponding to a power dispatching control command is input, identity verification is required, only the personnel with the power dispatching authority can perform subsequent voice control operation, irrelevant personnel are prevented from controlling the power system dispatching, the control reliability and safety are improved, and the voiceprints are similar to fingerprints and have uniqueness, and different people have different voiceprints, so that the security of identity verification can be improved by adopting the voiceprints to perform identity verification; identifying the voice data corresponding to each power dispatching control instruction to obtain corresponding text data, and extracting the characteristics to obtain corresponding characteristic data, comparing the text data with a preset special dictionary for the power dispatching control command, if the text data belongs to a word or a sentence in the special dictionary for the power dispatching control command, indicating that the text recognition is correct, retaining the text data, the text data is target text data, if the text data does not belong to words and sentences in the special power dispatching control instruction dictionary and indicates that the text recognition is wrong, the words and sentences which are the same as the pinyin of the recognized text data are searched from the special power dispatching control instruction dictionary and are replaced by the text data, the word is the target text data, a corresponding relation between the characteristic data and the corresponding target text data is finally established, and the power dispatching identity verification database and the corresponding relation form a power system dispatching control voice recognition model. The electric power dispatching control instruction is subjected to voice recognition and comparison by adopting the electric power dispatching control instruction special dictionary, the corresponding electric power dispatching control instruction can be finally obtained according to voice recognition, the voice recognition model obtained through the process is specially used for electric power system dispatching, the application of the voice recognition model in the electric power system dispatching is realized, the voice recognition accuracy and reliability in the electric power system dispatching field can be improved, and the requirement of electric power system dispatching with higher and higher intelligent requirements can be met.
Drawings
In order to more clearly illustrate the technical solution of the embodiment of the present invention, the drawings needed to be used in the embodiment will be briefly described as follows:
fig. 1 is a flow chart diagram of a power system scheduling control speech recognition model building method.
Detailed Description
As shown in fig. 1, the present embodiment provides a method for establishing a speech recognition model for power system dispatch control, where the method for establishing a speech recognition model for power system dispatch control can be used for speech recognition of a power system dispatch control instruction to implement power dispatch control. The execution main body of the power system dispatching control voice recognition model building method can be a computer, a server, an intelligent mobile terminal and the like.
The method for establishing the power system dispatching control voice recognition model comprises three parts, namely an establishing process of an identity verification part, an establishing process of a voice recognition part and an establishing process of the power system dispatching control voice recognition model. The establishing process of the identity authentication part and the establishing process of the voice recognition part are not in strict sequence, the establishing process of the identity authentication part can be carried out firstly, then the establishing process of the voice recognition part can be carried out, the establishing process of the voice recognition part can also be carried out firstly, then the establishing process of the identity authentication part can be carried out, and the establishing process of the identity authentication part and the establishing process of the voice recognition part can also be carried out simultaneously.
The establishment process of the identity authentication part comprises the following steps:
acquiring audio signals of personnel with power dispatching authority:
audio signals are captured by audio capture devices, which are many and in practical applications are named as: microphones, MICs (microphones), pickups, microphones, and so forth.
The object of the acquired audio signal is a person having power dispatching authority, such as: the staff who is specially responsible for the power scheduling, and high-level management personnel. The number of the personnel with the power dispatching authority is set according to actual needs. Therefore, the audio signals of the persons with the power dispatching authority are obtained through the related audio acquisition equipment.
Acquiring the voiceprint of the person corresponding to the audio signal according to the audio signal:
and after receiving the audio signal of any person with the power dispatching authority, extracting the voiceprint of the audio signal. Voiceprint (Voiceprint) is a spectrum of sound waves carrying verbal information. The voiceprint is unique like a fingerprint and has the function of identity recognition (identification of an individual). Each person has a specific voiceprint, which varies from person to person. Regardless of how one intentionally simulates the voice and tone of another, even if the simulation is vivid, the voiceprint is still different.
In order to improve the identification precision, when the audio signal of each person with the power dispatching authority is acquired, each person with the power dispatching authority reads the same section of characters or a word, so that the acquired audio signal of each person with the power dispatching authority is the same, and in the subsequent identification process, each identified object also reads the characters which are the same as the audio signals. Furthermore, the audio signal may be some more representative words or phrases, such as: "identify", "please authenticate", etc., or some specific number.
Then, the voice print of each person with power dispatching authority can be obtained by processing the audio signal of the person. Since the voiceprint extraction according to the audio signal belongs to the conventional technology, the details are not repeated.
Establishing a power dispatching identity verification database according to voiceprints of personnel with power dispatching authority:
and establishing a power dispatching identity verification database, wherein the power dispatching identity verification database comprises the obtained voiceprints of all the persons with the power dispatching authority. Then, in the subsequent control process (namely, the power system dispatching control voice recognition model obtained by the power system dispatching control voice recognition model establishing method is used for carrying out power dispatching control), acquiring an audio signal of an identified object, extracting a voiceprint, then comparing the obtained voiceprint with the power dispatching identity verification database, if the matching degree of the voiceprint and one voiceprint in the power dispatching identity verification database is greater than or equal to a preset value, indicating that the two voiceprints are the same voiceprint, wherein the voiceprint belongs to one voiceprint in the power dispatching identity verification database, and the identity of the identified object meets the requirement; if the matching degree of the voiceprint and each voiceprint in the power dispatching identity verification database is smaller than a preset value, the voiceprint is not matched with all voiceprints in the power dispatching identity verification database, the voiceprint does not belong to any voiceprint in the power dispatching identity verification database, and the identity of the identified object does not meet the requirement.
Further, after the step of "obtaining the voiceprint of the person corresponding to the audio signal", the establishing process of the authentication part further includes: and acquiring face images of personnel with power dispatching authority, and establishing corresponding relations between different voiceprints and corresponding face images. The face images of the personnel with the power dispatching authority are collected through the camera, and then the corresponding relation between the voiceprint of the personnel with the power dispatching authority and the corresponding face images is established.
Correspondingly, establishing a power dispatching identity verification database according to voiceprints of personnel with power dispatching authority includes: and establishing a power dispatching identity verification database according to the voiceprint and the corresponding face image of each person with the power dispatching authority, wherein the power dispatching identity verification database comprises the voiceprint and the corresponding face image of each person with the power dispatching authority.
Adding a face image on the basis of the voiceprint, in the subsequent control process, acquiring an audio signal and a face image of an identified object, extracting the voiceprint according to the audio signal, then comparing the obtained voiceprint with the power dispatching identity verification database, judging that the identity of the identified object preliminarily meets the requirement when the comparison meets the requirement, then comparing the face image of the identified object with the face image corresponding to the voiceprint of the identified object in the power dispatching identity verification database, and finally determining that the identity meets the requirement after the comparison of the face images passes. The identification precision can be improved by the identification mode.
The establishing process of the voice recognition part comprises the following steps:
acquiring voice data corresponding to each power dispatching control instruction:
in the establishing process of the voice recognition part, the number of the power dispatching control instructions is determined by actual needs, such as: if the electric power system used by the electric power system dispatching control voice recognition model establishing method provided by the embodiment is relatively small in scale, the required electric power dispatching control instructions are relatively few; accordingly, if the electric power system used by the electric power system scheduling control speech recognition model establishing method provided by the embodiment is large in scale, the required electric power scheduling control instructions are more. As a specific embodiment, the power scheduling control instruction may include: switch on, switch off, switch full phase trip, switch single phase trip, etc.
And acquiring voice data corresponding to each power dispatching control instruction, wherein each voice data corresponds to each power dispatching control instruction. If the power dispatching control command comprises: closing the switch, disconnecting the switch and tripping the switch in a full phase, wherein the corresponding voice data is as follows: the method comprises the steps of switching on a switch, switching off the switch and switching full-phase tripping. It should be appreciated that, in order to improve the accuracy of the subsequent text recognition, the relevant personnel who are more skilled in the mandarin can read each power scheduling control instruction to obtain the corresponding voice data.
Performing feature extraction on the voice data to obtain corresponding feature data:
before feature extraction is carried out on voice data corresponding to each power dispatching control instruction, denoising processing can be carried out on the voice data, interference signals such as noise in the voice data are removed, and accuracy of feature extraction is improved.
Different voice data have certain characteristics, the characteristics refer to data information indicating the specificity of the voice data, and different voice data have different characteristic data. Then, feature extraction is performed on each voice data, and corresponding feature data can be obtained. The implementation process of extracting features of voice data to obtain corresponding feature data belongs to conventional technical means, and is not described in detail.
The feature data is used for subsequent control (namely, the power system dispatching control speech recognition model obtained by the power system dispatching control speech recognition model establishing method is used for power dispatching control), after the speech data of the identified object is obtained, the feature data of the speech data needs to be extracted, and then the corresponding power dispatching control instruction is obtained through identification according to the feature data and the established power system dispatching control speech recognition model.
Recognizing the voice data to obtain corresponding text data:
and identifying the voice data corresponding to each power dispatching control instruction to obtain text data of the voice data corresponding to each power dispatching control instruction. I.e. speech data is recognized as text data. Such as recognizing the voice data "on switch" as the text data "on switch". Of course, before the speech data is recognized to obtain the corresponding text data, denoising processing may be performed on the speech data, and then text recognition may be performed to obtain the text data. Because the text recognition is carried out on the voice data, the corresponding text data obtained belongs to the conventional technical means, and is not described in detail.
Comparing the text data with a preset special power dispatching control instruction dictionary, and if the text data belongs to words and sentences in the special power dispatching control instruction dictionary, retaining the text data to obtain target text data; if the text data does not belong to the words and phrases in the electric power dispatching control instruction special dictionary, searching the words and phrases in the electric power dispatching control instruction special dictionary which are the same as the pinyin of the text data according to the pinyin of the text data, replacing the text data with the searched words and phrases to obtain target text data:
when the voice data is recognized as text data, a recognition error may occur, where the recognition error refers to a different text recognized as the same pinyin (i.e., the same pronunciation), and the obtained text is not a proper noun in the field of power scheduling control, i.e., is not a power scheduling control command. For example, if "on" in the voice data "on switch" is recognized as text data "cheer" and "on" in the voice data "on switch" is recognized as text data "open hall", then the voice data "on switch" is recognized as text data "cheer open hall". It is obvious that "shang chang" is not a power dispatching control command.
In order to avoid the recognition error, a power scheduling control instruction-specific dictionary is preset, and words corresponding to each power scheduling control instruction, that is, text data corresponding to each power scheduling control instruction, are stored in the power scheduling control instruction-specific dictionary. It should be understood that, the amount of text data corresponding to the power dispatching control instruction contained in the power dispatching control instruction special dictionary is set according to actual needs, as described above, if the power system used in the power system dispatching control voice recognition model building method provided in this embodiment is relatively small in scale, the amount of text data corresponding to the power dispatching control instruction contained in the power dispatching control instruction special dictionary may be relatively small; accordingly, if the electric power system used in the electric power system dispatching control voice recognition model establishing method provided in this embodiment is relatively large in scale, more text data corresponding to the electric power dispatching control command included in the electric power dispatching control command special dictionary may be set.
Then, comparing each obtained text data with a preset power dispatching control instruction special dictionary, and taking one of the text data as an example for explanation: the text data is compared with a dictionary special for the power dispatching control instruction, specifically, the text data is compared with each word and sentence in the dictionary special for the power dispatching control instruction one by one, the comparison mode can be an automatic comparison mode, the text data is automatically compared with each word and sentence in the dictionary special for the power dispatching control instruction one by one according to a computer program, the comparison mode can also be a manual comparison mode, namely, each word and sentence in the dictionary special for the power dispatching control instruction is placed in a document, and whether the text data exists in the document is searched by adopting a keyword searching mode.
And if the text data belongs to words and sentences in the special dictionary for the power dispatching control instruction, reserving the text data and taking the text data as target text data.
If the text data does not belong to the words and phrases in the electric power dispatching control instruction special dictionary, the words and phrases which are the same as the pinyin of the text data are searched from the electric power dispatching control instruction special dictionary according to the pinyin of the text data, the searched words and phrases are used for replacing the text data, and the searched words and phrases are the target text data. In order to implement the above replacement process, the power scheduling control instruction dedicated dictionary includes words and sentences of each power scheduling control instruction and pinyin of the words and sentences of each power scheduling control instruction (since accurate spelling of voice tones cannot be absolutely guaranteed, the pinyin may not include tones), that is, the power scheduling control instruction dedicated dictionary includes the pinyin of each word and sentence in addition to each word and sentence. Accordingly, a specific implementation is given below:
firstly, obtaining pinyin of the text data;
then, inputting the obtained pinyin of the text data into a special power dispatching control instruction dictionary to obtain the pinyin of the power dispatching control instruction dictionary, wherein the pinyin of the text data is the same as the pinyin of the text data, specifically, the pinyin of the text data is compared with the pinyin of each word and phrase in the special power dispatching control instruction dictionary one by one to find the pinyin of the power dispatching control instruction dictionary, which is the same as the pinyin of the text data, and then, the corresponding word and phrase in the special power dispatching control instruction dictionary is obtained according to the obtained pinyin, wherein the word and phrase is the required word and phrase;
and finally, replacing the text data with the searched words and sentences, wherein the searched words and sentences are the target text data. Such as: if the text data is ' opening a hall with the sum ', and is not a word or a sentence in the special dictionary for the power dispatching control instruction, acquiring the pinyin of the text data ' opening a hall with the sum ', the he shang kai guan, comparing the pinyin with the pinyin of each word or sentence in the special dictionary for the power dispatching control instruction one by one, finding the same pinyin ' heshang kang kai guan ' in the special dictionary for the power dispatching control instruction, acquiring a word or a sentence ' closing switch ' corresponding to the pinyin ' he shang kai guan ', and replacing the text data ' and opening the hall with the text data ' closing switch ', wherein the text data ' closing switch ' is the target text data.
Through the above process, the target text data corresponding to each voice data can be obtained.
Establishing a corresponding relation between each feature data and corresponding target text data:
since the feature data of each voice data and the target text data of each voice data are obtained as described above, the corresponding relationship between each feature data and the corresponding target text data is established.
The establishing process of the identity verification part obtains a power dispatching identity verification database, the establishing process of the voice recognition part obtains the corresponding relation between each characteristic data and the corresponding target text data, and then the corresponding relation between the power dispatching identity verification database and each characteristic data and the corresponding target text data forms a power system dispatching control voice recognition model.
When the power system dispatching control voice recognition model is used for carrying out power dispatching control, acquiring an audio signal of a recognized object, extracting voiceprints, comparing the obtained voiceprints with a power dispatching identity verification database in the power system dispatching control voice recognition model, and if the matching degree of the voiceprints and one voiceprint in the power dispatching identity verification database is larger than or equal to a preset value, indicating that the two voiceprints are the same voiceprint, wherein the voiceprint belongs to one voiceprint in the power dispatching identity verification database, and the identity of the recognized object meets the requirement; if the matching degree of the voiceprint and each voiceprint in the power dispatching identity verification database is smaller than a preset value, the voiceprint is not matched with all voiceprints in the power dispatching identity verification database, the voiceprint does not belong to any voiceprint in the power dispatching identity verification database, and the identity of the identified object does not meet the requirement.
If the identity of the recognized object meets the requirement, voice data of the recognized object is obtained, then feature extraction is carried out on the voice data to obtain corresponding feature data, target text data corresponding to the feature data are determined according to the corresponding relation between each feature data and the corresponding target text data in the power system dispatching control voice recognition model, and then a corresponding power dispatching control instruction is obtained.
The above-mentioned embodiments are merely illustrative of the technical solutions of the present invention in a specific embodiment, and any equivalent substitutions and modifications or partial substitutions of the present invention without departing from the spirit and scope of the present invention should be covered by the claims of the present invention.
Claims (3)
1. A power system dispatching control voice recognition model building method is characterized by comprising the following steps:
acquiring audio signals of personnel with power dispatching authority;
acquiring voiceprints of people corresponding to the audio signals according to the audio signals;
establishing a power dispatching identity verification database according to voiceprints of personnel with power dispatching authority;
acquiring voice data corresponding to each power dispatching control instruction;
extracting the characteristics of the voice data to obtain corresponding characteristic data;
recognizing the voice data to obtain corresponding text data;
comparing the text data with a preset special power dispatching control instruction dictionary, and if the text data belongs to words and sentences in the special power dispatching control instruction dictionary, retaining the text data to obtain target text data; if the text data does not belong to words and sentences in the electric power dispatching control instruction special dictionary, searching words and sentences in the electric power dispatching control instruction special dictionary which are the same as the pinyin of the text data according to the pinyin of the text data, and replacing the text data with the searched words and sentences to obtain target text data;
establishing a corresponding relation between each characteristic data and corresponding target text data;
and the power dispatching identity verification database and the corresponding relation form a power system dispatching control voice recognition model.
2. The method for establishing the speech recognition model for dispatching and controlling the power system according to claim 1, wherein after acquiring the voiceprint of the person corresponding to the audio signal, the method for establishing the speech recognition model for dispatching and controlling the power system further comprises:
acquiring face images of personnel with power dispatching authority, and establishing corresponding relations between different voiceprints and corresponding face images;
correspondingly, the establishing of the power dispatching identity verification database according to the voiceprints of the personnel with the power dispatching authority comprises the following steps:
and establishing a power dispatching identity verification database according to the voiceprint and the corresponding face image of each person with the power dispatching authority, wherein the power dispatching identity verification database comprises the voiceprint and the corresponding face image of each person with the power dispatching authority.
3. The power system scheduling control speech recognition model building method according to claim 1, wherein the power scheduling control instruction dedicated dictionary includes words of each power scheduling control instruction and pinyin of the words of each power scheduling control instruction;
the searching for the word and sentence in the power scheduling control instruction special dictionary which is the same as the pinyin of the text data according to the pinyin of the text data, and replacing the searched word and sentence with the text data to obtain target text data comprises:
obtaining pinyin of the text data;
inputting the obtained pinyin of the text data into the power dispatching control instruction special dictionary to obtain the pinyin which is the same as the pinyin of the text data in the power dispatching control instruction special dictionary, and obtaining corresponding words and sentences in the power dispatching control instruction special dictionary according to the obtained pinyin;
and replacing the text data with the obtained words and sentences to obtain target text data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010299229.5A CN111462754B (en) | 2020-04-16 | 2020-04-16 | Method for establishing dispatching control voice recognition model of power system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010299229.5A CN111462754B (en) | 2020-04-16 | 2020-04-16 | Method for establishing dispatching control voice recognition model of power system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111462754A true CN111462754A (en) | 2020-07-28 |
| CN111462754B CN111462754B (en) | 2022-08-09 |
Family
ID=71681246
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010299229.5A Active CN111462754B (en) | 2020-04-16 | 2020-04-16 | Method for establishing dispatching control voice recognition model of power system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111462754B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112651854A (en) * | 2020-12-23 | 2021-04-13 | 讯飞智元信息科技有限公司 | Voice scheduling method and device, electronic equipment and storage medium |
| CN113283695A (en) * | 2021-04-19 | 2021-08-20 | 贵州电网有限责任公司 | Power dispatching intelligent agent implementation method and system based on artificial intelligence |
| CN116013291A (en) * | 2022-12-12 | 2023-04-25 | 广西电网有限责任公司 | Speech recognition system based on ontology knowledge in electric power field |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102780706A (en) * | 2012-08-14 | 2012-11-14 | 俞琳 | User authentication method based on power system service |
| CN105810200A (en) * | 2016-02-04 | 2016-07-27 | 深圳前海勇艺达机器人有限公司 | Man-machine dialogue apparatus and method based on voiceprint identification |
| CN108121528A (en) * | 2017-12-06 | 2018-06-05 | 深圳市欧瑞博科技有限公司 | Sound control method, device, server and computer readable storage medium |
| CN108199480A (en) * | 2017-12-27 | 2018-06-22 | 国电南瑞南京控制系统有限公司 | A kind of electric power dispatching system and its method based on artificial intelligence |
| CN109559737A (en) * | 2018-12-13 | 2019-04-02 | 朱明增 | Electric power system dispatching speech model method for building up |
| CN110473566A (en) * | 2019-07-25 | 2019-11-19 | 深圳壹账通智能科技有限公司 | Audio separation method, device, electronic equipment and computer readable storage medium |
| CN110728973A (en) * | 2019-10-23 | 2020-01-24 | 青岛聚看云科技有限公司 | Video resource output method and server |
-
2020
- 2020-04-16 CN CN202010299229.5A patent/CN111462754B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102780706A (en) * | 2012-08-14 | 2012-11-14 | 俞琳 | User authentication method based on power system service |
| CN105810200A (en) * | 2016-02-04 | 2016-07-27 | 深圳前海勇艺达机器人有限公司 | Man-machine dialogue apparatus and method based on voiceprint identification |
| CN108121528A (en) * | 2017-12-06 | 2018-06-05 | 深圳市欧瑞博科技有限公司 | Sound control method, device, server and computer readable storage medium |
| CN108199480A (en) * | 2017-12-27 | 2018-06-22 | 国电南瑞南京控制系统有限公司 | A kind of electric power dispatching system and its method based on artificial intelligence |
| CN109559737A (en) * | 2018-12-13 | 2019-04-02 | 朱明增 | Electric power system dispatching speech model method for building up |
| CN110473566A (en) * | 2019-07-25 | 2019-11-19 | 深圳壹账通智能科技有限公司 | Audio separation method, device, electronic equipment and computer readable storage medium |
| CN110728973A (en) * | 2019-10-23 | 2020-01-24 | 青岛聚看云科技有限公司 | Video resource output method and server |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112651854A (en) * | 2020-12-23 | 2021-04-13 | 讯飞智元信息科技有限公司 | Voice scheduling method and device, electronic equipment and storage medium |
| CN113283695A (en) * | 2021-04-19 | 2021-08-20 | 贵州电网有限责任公司 | Power dispatching intelligent agent implementation method and system based on artificial intelligence |
| CN116013291A (en) * | 2022-12-12 | 2023-04-25 | 广西电网有限责任公司 | Speech recognition system based on ontology knowledge in electric power field |
| CN116013291B (en) * | 2022-12-12 | 2024-05-14 | 广西电网有限责任公司 | Speech recognition system based on ontology knowledge in electric power field |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111462754B (en) | 2022-08-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108304372B (en) | Entity extraction method and device, computer equipment and storage medium | |
| US10476872B2 (en) | Joint speaker authentication and key phrase identification | |
| CN111462754B (en) | Method for establishing dispatching control voice recognition model of power system | |
| CN101076851B (en) | Spoken language identification system and method for training and operating the said system | |
| CN109192194A (en) | Voice data mask method, device, computer equipment and storage medium | |
| CN111324708A (en) | Natural language processing system based on human-computer interaction | |
| CN113535925B (en) | Voice broadcasting method, device, equipment and storage medium | |
| CN111402892A (en) | Conference recording template generation method based on voice recognition | |
| CN106782503A (en) | Automatic speech recognition method based on physiologic information in phonation | |
| WO2021118072A1 (en) | Method and device for detecting significant section by using audio and video information | |
| CN113409774A (en) | Voice recognition method and device and electronic equipment | |
| CN113436617B (en) | Voice sentence breaking method, device, computer equipment and storage medium | |
| CN113051384A (en) | User portrait extraction method based on conversation and related device | |
| KR20110065916A (en) | Interpretation system for error correction and auto scheduling | |
| CN112863485A (en) | Accent voice recognition method, apparatus, device and storage medium | |
| CN110838290A (en) | Voice robot interaction method and device for cross-language communication | |
| CN111276146A (en) | Teaching training system based on voice recognition | |
| CN115547345A (en) | Voiceprint recognition model training and related recognition method, electronic device and storage medium | |
| CN114974310A (en) | Emotion recognition method and device based on artificial intelligence, computer equipment and medium | |
| CN114203160A (en) | Method, device and equipment for generating sample data set | |
| JP2010197709A (en) | Voice recognition response method, voice recognition response system and program therefore | |
| WO2020096073A1 (en) | Method and device for generating optimal language model using big data | |
| CN101304457A (en) | Method and apparatus for implementing automatic spoken language training based on voice telephone | |
| CN117995167B (en) | Japanese speech translation method based on machine translation, terminal and storage medium | |
| CN117789706B (en) | Audio information content identification method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |