CN107017007A - A kind of substation field operation remote command method based on voice transfer - Google Patents
A kind of substation field operation remote command method based on voice transfer Download PDFInfo
- Publication number
- CN107017007A CN107017007A CN201710334346.9A CN201710334346A CN107017007A CN 107017007 A CN107017007 A CN 107017007A CN 201710334346 A CN201710334346 A CN 201710334346A CN 107017007 A CN107017007 A CN 107017007A
- Authority
- CN
- China
- Prior art keywords
- voice
- sub
- data
- sound
- parameter
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000012546 transfer Methods 0.000 title claims abstract description 30
- 230000015572 biosynthetic process Effects 0.000 claims abstract description 18
- 238000003786 synthesis reaction Methods 0.000 claims abstract description 18
- 230000008569 process Effects 0.000 claims abstract description 17
- 230000005540 biological transmission Effects 0.000 claims abstract description 14
- 238000011282 treatment Methods 0.000 claims abstract description 8
- 230000002452 interceptive effect Effects 0.000 claims abstract description 5
- 230000011664 signaling Effects 0.000 claims description 26
- 238000012360 testing method Methods 0.000 claims description 22
- 238000012549 training Methods 0.000 claims description 22
- 238000010276 construction Methods 0.000 claims description 16
- 238000005086 pumping Methods 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 12
- 238000013139 quantization Methods 0.000 claims description 10
- 239000000203 mixture Substances 0.000 claims description 7
- 230000000630 rising effect Effects 0.000 claims description 6
- 230000001360 synchronised effect Effects 0.000 claims description 6
- 230000006835 compression Effects 0.000 claims description 3
- 238000007906 compression Methods 0.000 claims description 3
- 239000012141 concentrate Substances 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 238000009432 framing Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 230000004044 response Effects 0.000 claims description 3
- 230000005284 excitation Effects 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 5
- 238000011002 quantification Methods 0.000 abstract description 4
- 238000004891 communication Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 241000208340 Araliaceae Species 0.000 description 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 2
- 235000003140 Panax quinquefolius Nutrition 0.000 description 2
- 230000004888 barrier function Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 235000008434 ginseng Nutrition 0.000 description 2
- 238000002955 isolation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000035939 shock Effects 0.000 description 2
- 241000196324 Embryophyta Species 0.000 description 1
- 230000000712 assembly Effects 0.000 description 1
- 238000000429 assembly Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000009792 diffusion process Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/93—Discriminating between voiced and unvoiced parts of speech signals
-
- 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
- G10L19/00—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis
- G10L19/02—Speech or audio signals analysis-synthesis techniques for redundancy reduction, e.g. in vocoders; Coding or decoding of speech or audio signals, using source filter models or psychoacoustic analysis using spectral analysis, e.g. transform vocoders or subband vocoders
- G10L19/032—Quantisation or dequantisation of spectral components
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04Q—SELECTING
- H04Q5/00—Selecting arrangements wherein two or more subscriber stations are connected by the same line to the exchange
- H04Q5/24—Selecting arrangements wherein two or more subscriber stations are connected by the same line to the exchange for two-party-line systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Computational Linguistics (AREA)
- Signal Processing (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Human Computer Interaction (AREA)
- Acoustics & Sound (AREA)
- Multimedia (AREA)
- Computer Networks & Wireless Communication (AREA)
- Telephonic Communication Services (AREA)
Abstract
The invention discloses a kind of substation field operation remote command method based on voice transfer, it decodes three processes by encoded speech data, data transfer and speech data and carries out remote speech transmission, with the interactive voice of this commander's talk back equipment for completing live talk back equipment and remote command center, so that to instruct substation field operation, code efficiency is effectively raised, and obtains more preferable sense of hearing effect；And the pure and impure tone parameter of subband has carried out quantification treatment in sound transmission course, improves the quantitative efficiency of sub-band surd and sonant degree parameter, the naturalness of synthesis voice is improved.
Description
Technical field
The present invention relates to a kind of substation field operation remote command method based on voice transfer, belong to substation safety
Operating technology field.
Background technology
It is directly used in production and uses electric energy, the electrical equipment higher than control loop (secondary device) voltage is referred to as once set
It is standby.Main composition has：Generator (motor), transformer, breaker, disconnecting switch, automatic switch, contactor, knife-like switch, mother
Line, transmission line of electricity, power cable, reactor etc..Operation for primary equipment needs high degree of safety,《Electric power safety work
Make code》Middle regulation, the operation of operations staff is divided into monitoring operation and one man operation.Wherein, monitoring operation is held by least two people
OK, wherein a people guards to the equipment person of being familiar with, one man operation is by people's complete independently.At present in operation, one
As in the case of, administrative staff can not know the concrete condition of operation, and when operation goes wrong, administrative staff at most can only
Gone and found out what's going on by phone, it is impossible to obtain live visual information.The drawbacks of this operational method has certain, particularly operator
Member is to some primary equipment when will operate, administrative staff control centre miscellaneous equipment is turned off or change its course so that pair
The primary equipment can be with safety operation, but administrative staff extremely easily select primary equipment to malfunction, will it is other once
Equipment safety is isolated, and the primary equipment that will be operated is not isolated from, and causes very big potential safety hazard, or even harm execute-in-place people
The life of member.This occurred many in the actual course of work, but prevented this feelings there is presently no preferable method
The generation of condition, sees clearly equipment to isolate in can only working again, dependence be staff careful degree.
The utility model of Application No. 201120347568.2 discloses a kind of transforming plant protecting, observing and controlling screen cabinet voice message
Device, including voice module, power supply, horn module, electromagnetism normally closed contact and electromagnet, wherein electromagnetism normally closed contact are arranged on guarantor
In shield, the doorframe of observing and controlling screen cabinet, electromagnet is arranged in cabinet door, one end of electromagnetism normally closed contact and the triggering of voice cue module
Signal input part is connected, and other end ground connection, the voice signal output end of voice module and the signal input part of loudspeaker are connected, power supply
Powered to modules.Operator reports the device name to be operated with voice, kept away when opening protection, observing and controlling screen cabinet door
Exempt from operator and go to the wrong way interval to cause maloperation, prevent the generation of artificial misoperation fault from source, improve electric network reliability.
The utility model is once to be thought verification before operation, but people can have fault phenomenon, it is possible to causing meaning
Outer accident.
During remote command, speech quality is to determine a key factor of field operation success or failure, speech quality
Quality influence both sides' communication effectiveness and operation quality.Accordingly, it would be desirable to a kind of high-quality voice transmission method.
The content of the invention
It is existing the invention provides a kind of transformer station based on voice transfer in order to overcome the shortcomings of that above-mentioned prior art is present
Field operation remote command method, it is checked the primary equipment at scene by voice dialogue, makes administrative staff's remote command
Site operation personnel carries out safety operation.
The technical scheme adopted by the invention to solve the technical problem is that：A kind of substation field based on voice transfer is made
Industry remote command method, carries out interactive voice to instruct by commander's talk back equipment of live talk back equipment and remote command center
Substation field operation, it is characterized in that, including procedure below：
First, encoded speech data
Step 11, obtain live talk back equipment or command the speech data of the telephone transmitter input of talk back equipment；
Step 12, the speech data of acquisition is subjected to SILK coded treatments and generates voice sub-band parameter；
Step 13, voice sub-band surd and sonant parameter is quantified；The step 13 is comprised the following steps that：
Step 131, sub-band surd and sonant in voice sub-band parameter is judged；Subband is clear in the sub-band parameter to voice
The process that voiced sound is judged is：Pure and impure sound is set to judge data set first；Then voice training data are utilized and described pure and impure
Sound judges data set, and the pure and impure sound decision model for judging tree construction to y-bend is trained, and wherein the y-bend is judged in tree construction
Non-leaf nodes judge the problem of data are concentrated for the pure and impure sound, leaf node is pure and impure sound result of determination；Finally receive
Tone testing data, and tone testing data are judged as voiceless sound or voiced sound using the pure and impure sound decision model after training, when
Tone testing data are judged as after voiced sound, and the pumping signal of the tone testing data is assumed to be during phonetic synthesis
Shock response sequence；After tone testing data are judged as voiceless sound, the tone testing data during phonetic synthesis
Pumping signal is assumed to be white noise；
Step 132, timer m1=0 is exited in coding side setting, and interval T0 is exited in setting；
Step 133, to the speech data sampling point framing in chronological order of acquisition, and sub-band surd and sonant degree ginseng is extracted by frame
Number；
Step 134, multiple speech frame composition superframes, sub-band surd and sonant degree parameter composition superframe of multiple speech frames are combined
With pure and impure tone parameter；
Step 135, judge m1 whether the integral multiple for being T0, if it is using initial codebook to superframe sub-band surd and sonant degree
Parameter carries out vector quantization；Otherwise the sub-band surd and sonant degree parameter after a upper superframe is quantified is quantified using initial codebook,
Assuming that quantized value is n, the corresponding code book Cn of the index is chosen from condition codebook set according to quantization index value n, code book is utilized
Cn is quantified to current super frame sub-band surd and sonant degree parameter and obtains index value index；Described initial codebook and condition code
This collection is obtained by off-line training；
Step 136, index value is subjected to coding transmission, makes m1=m1+1 and go to step 135；
Step 14, by the corresponding bit stream of voice sub-band parameter compression generation current speech data after real-time quantization；
2nd, data transfer
Live talk back equipment or the speech data for the telephone transmitter input for commanding talk back equipment are sent to commander's talk back equipment
Or the telephone receiver of live talk back equipment；
The detailed process of the data transfer comprises the following steps：
Step 21, telephone transmitter send voice superframes when, in the head time slot of each voice superframes place synchronous code with it is complete
Whole signaling parameter, the signaling parameter includes all parameters for being used to carry out the foundation of voice call link；
Step 22, when telephone receiver receives signal, detect the signal that receives whether be voice superframes voice head time slot, such as
Fruit is then to be transferred to step 23, otherwise, is transferred to step 24；
Step 23, the signaling parameter of carrying is parsed from the voice head time slot, and is entered according to the signaling parameter parsed
The foundation of row voice call link；
Step 24, according to the content for receiving signal, signal transacting is carried out；The concrete operations of the step 24 are：If connect
The signal received is data head time slot, then sets up data interaction link；If the signal received is data block pdu, skill
According to block；If the signal received is high-rise control signaling, the high-rise control signaling of parsing；
3rd, speech data is decoded
Step 31, the corresponding bit stream of current speech data is obtained, and by SILK decoders to current speech data ratio
Spy's stream decodes the voice sub-band parameter after being quantified；
Step 32, extract quantify after voice sub-band parameter each sub-band surd and sonant parameter and synthesize pumping signal；The step
Rapid 32 comprise the following steps that：
Step 321, decoding end exits timer m2=0 with coding side setting simultaneously, and interval T0 is exited in setting；
Step 322, decoding end judge m2 whether the integral multiple for being T0, if it is search for initial according to index value index
Code book obtains superframe sub-band surd and sonant degree parameter；Otherwise the sub-band surd and sonant degree after being quantified using initial codebook to a upper superframe is joined
Number is quantified, and obtains index value n, and concentrates selection code book Cn from condition code book according to index value n, is received according to decoding end
Index value index search code book Cn obtains superframe sub-band surd and sonant degree parameter；
Step 323, according to superframe sub-band surd and sonant degree parameter, the sub-band surd and sonant degree parameter of each speech frame is obtained, is sent
Enter pumping signal synthesis end generation pumping signal；Make m=m+1 and go to step 322；
Step 33, the voice of synthesis is exported in real time.
Preferably, the process that the pure and impure sound decision model for judging y-bend tree construction is trained as：For pure and impure
Sound judges each problem that data are concentrated, and is to calculate voiced sound ratio value respectively with no voice training data for answering, and
And it is used as root node the problem of selection voiced sound proportional difference maximum；Enter line splitting for voice training data under root node, with
Form non-leaf nodes and leaf node.
Preferably, during the pure and impure sound decision model for judging y-bend tree construction is trained,
After division stop condition set in advance is met, stop the division, wherein the division stops bar
Part is：Non-leaf nodes or the voice training data of leaf node are less than the first threshold value set in advance；
After division stop condition set in advance is met, stop the division, wherein the division stops bar
Part is：The voiced sound proportional difference of non-leaf nodes or leaf node is less than the second threshold value set in advance.
Preferably, the pure and impure sound judges whether data set is vowel, plosive, friction as the phoneme belonging to speech frame
Sound, read again, nasal sound, high and level tone, rising tone, upper sound or falling tone, and forward and backward one in speech sentence of phoneme belonging to speech frame
Whether phoneme is vowel, plosive, fricative, stressed, nasal sound, high and level tone, rising tone, upper sound or falling tone.
Preferably, interval T0 is exited to be set according to the size of superframe.
Preferably, the speech data sampling point to acquisition is by 8kHz frequency samplings and done by high-pass filtering removal power frequency
The voice sampling point disturbed.
Preferably, the quantity y of the condition code book and the quantizing bit number x of sub-band surd and sonant degree parameter relational expression are：y
=2x。
The beneficial effects of the invention are as follows,
The present invention decodes three processes by encoded speech data, data transfer and speech data and carries out remote speech biography
It is defeated, with the interactive voice of this commander's talk back equipment for completing live talk back equipment and remote command center, so as to instruct power transformation
Stand field operation, effectively raise code efficiency, and obtain more preferable sense of hearing effect.
Checked by the command methods of the present invention come the primary equipment to substation field operation, it is ensured that to correct
Primary equipment is isolated, and verification is checked using flexecutive's real-time instruction, it is ensured that the accuracy of isolation；The present invention
Convenient scene and the communication of control centre, can be sent to control centre by the sound at scene, be easy to the technical barrier to scene
Quickly solved, administrative staff's remote command site operation personnel is carried out safety operation.
Present invention subband pure and impure tone parameter in sound transmission course has carried out quantification treatment, it is contemplated that continuous superframe
The correlation of sub-band surd and sonant degree parameter, the principle of adoption status transfer during to current super frame sub-band surd and sonant metrization, is examined
Consider the sub-band surd and sonant degree parameter of a upper superframe, therefore improve the quantitative efficiency of sub-band surd and sonant degree parameter.The present invention
The naturalness of synthesis voice is improved, is adapted to low rate and ultralow speed parametric speech coding with below 2400bps.
The present invention during sub-band surd and sonant degree parameter quantification treatment first to voice sub-band parameter in sub-band surd and sonant
Judged, it sets pure and impure sound to judge data set；Using voice training data and the pure and impure sound decision problem collection, to y-bend
Judge that the pure and impure sound decision model of tree construction is trained, wherein the y-bend judges the non-leaf nodes in tree construction to be described clear
The problem of voiced sound decision problem is concentrated, leaf node is pure and impure sound result of determination；Tone testing data are received, and are utilized after training
The pure and impure sound decision model judge tone testing data as voiceless sound or voiced sound.As can be seen here, embodiment of the present invention is used
Independent pure and impure sound decision model pairing is judged into plain the pure and impure of state of voice middle pitch, is judged into so as to improve pure and impure sound
Power, solve synthesis voice voiced sound Qinghua and voiceless sound turbidity caused by the low defect of synthetic effect, improve language
The quality of sound synthesis.
The data transfer mode of the present invention can reduce overhead on the premise of system transmission performance is not reduced,
System efficiency of transmission is improved, and reduces system processing complexity, the flow of voice transfer link establishment is more rationalized.
Brief description of the drawings
The present invention is further described below in conjunction with the accompanying drawings：
Fig. 1 is flow chart of the method for the present invention；
Fig. 2 is the flow chart of encoded speech data of the present invention；
Fig. 3 is the flow chart of data transfer of the present invention；
Fig. 4 is the flow chart of speech data decoding of the present invention.
Embodiment
For the technical characterstic for illustrating this programme can be understood, below by embodiment, and its accompanying drawing is combined, to this hair
It is bright to be described in detail.Following disclosure provides many different embodiments or example is used for realizing the different knots of the present invention
Structure.In order to simplify disclosure of the invention, hereinafter the part and setting of specific examples are described.In addition, the present invention can be with
Repeat reference numerals and/or letter in different examples.This repetition is that for purposes of simplicity and clarity, itself is not indicated
Relation between various embodiments are discussed and/or set.It should be noted that part illustrated in the accompanying drawings is not necessarily to scale
Draw.Present invention omits the description to known assemblies and treatment technology and process to avoid being unnecessarily limiting the present invention.
As shown in figure 1, a kind of substation field operation remote command method based on voice transfer of the present invention, the present invention
Three processes are decoded by encoded speech data, data transfer and speech data and carry out remote speech transmission, scene is completed with this
The interactive voice of commander's talk back equipment of talk back equipment and remote command center, so as to instruct substation field operation, effectively
Improve code efficiency, and obtain more preferable sense of hearing effect., command methods, which are somebody's turn to do, include procedure below.
First, encoded speech data
As shown in Fig. 2 the detailed process of encoded speech data comprises the following steps：
Step 11, obtain live talk back equipment or command the speech data of the telephone transmitter input of talk back equipment；
Step 12, the speech data of acquisition is subjected to SILK coded treatments and generates voice sub-band parameter；
Step 13, voice sub-band surd and sonant parameter is quantified；The step 13 is comprised the following steps that：
Step 131, sub-band surd and sonant in voice sub-band parameter is judged；Subband is clear in the sub-band parameter to voice
The process that voiced sound is judged is：Pure and impure sound is set to judge data set first；Then voice training data are utilized and described pure and impure
Sound judges data set, and the pure and impure sound decision model for judging tree construction to y-bend is trained, and wherein the y-bend is judged in tree construction
Non-leaf nodes judge the problem of data are concentrated for the pure and impure sound, leaf node is pure and impure sound result of determination；Finally receive
Tone testing data, and tone testing data are judged as voiceless sound or voiced sound using the pure and impure sound decision model after training, when
Tone testing data are judged as after voiced sound, and the pumping signal of the tone testing data is assumed to be during phonetic synthesis
Shock response sequence；After tone testing data are judged as voiceless sound, the tone testing data during phonetic synthesis
Pumping signal is assumed to be white noise；
Step 132, timer m1=0 is exited in coding side setting, and interval T0 is exited in setting；
Step 133, to the speech data sampling point framing in chronological order of acquisition, and sub-band surd and sonant degree ginseng is extracted by frame
Number；
Step 134, multiple speech frame composition superframes, sub-band surd and sonant degree parameter composition superframe of multiple speech frames are combined
With pure and impure tone parameter；
Step 135, judge m1 whether the integral multiple for being T0, if it is using initial codebook to superframe sub-band surd and sonant degree
Parameter carries out vector quantization；Otherwise the sub-band surd and sonant degree parameter after a upper superframe is quantified is quantified using initial codebook,
Assuming that quantized value is n, the corresponding code book Cn of the index is chosen from condition codebook set according to quantization index value n, code book is utilized
Cn is quantified to current super frame sub-band surd and sonant degree parameter and obtains index value index；Described initial codebook and condition code
This collection is obtained by off-line training；
Step 136, index value is subjected to coding transmission, makes m1=m1+1 and go to step 135；
Step 14, by the corresponding bit stream of voice sub-band parameter compression generation current speech data after real-time quantization.
The present invention during sub-band surd and sonant degree parameter quantification treatment first to voice sub-band parameter in sub-band surd and sonant
Judged, it sets pure and impure sound to judge data set；Using voice training data and the pure and impure sound decision problem collection, to y-bend
Judge that the pure and impure sound decision model of tree construction is trained, wherein the y-bend judges the non-leaf nodes in tree construction to be described clear
The problem of voiced sound decision problem is concentrated, leaf node is pure and impure sound result of determination；Tone testing data are received, and are utilized after training
The pure and impure sound decision model judge tone testing data as voiceless sound or voiced sound.As can be seen here, embodiment of the present invention is used
Independent pure and impure sound decision model pairing is judged into plain the pure and impure of state of voice middle pitch, is judged into so as to improve pure and impure sound
Power, solve synthesis voice voiced sound Qinghua and voiceless sound turbidity caused by the low defect of synthetic effect, improve language
The quality of sound synthesis.
2nd, data transfer
Live talk back equipment or the speech data for the telephone transmitter input for commanding talk back equipment are sent to commander's talk back equipment
Or the telephone receiver of live talk back equipment.As shown in figure 3, the detailed process of data transfer comprises the following steps：
Step 21, telephone transmitter send voice superframes when, in the head time slot of each voice superframes place synchronous code with it is complete
Whole signaling parameter, the signaling parameter includes all parameters for being used to carry out the foundation of voice call link；Synchronous code is used for
Receiving terminal is synchronous with transmitting terminal；
Step 22, when telephone receiver receives signal, detect the signal that receives whether be voice superframes voice head time slot, such as
Fruit is to be transferred to step 23, is otherwise transferred to step 24；
Step 23, the signaling parameter of carrying is parsed from the voice head time slot, and is entered according to the signaling parameter parsed
The foundation of row voice call link；
Step 24, according to the content for receiving signal, signal transacting is carried out；The concrete operations of the step 24 are：If connect
The signal received is data head time slot, then sets up data interaction link；If the signal received is data block pdu, skill
According to block；If the signal received is high-rise control signaling, the high-rise control signaling of parsing.
Signaling parameter described above includes physical layer signaling and high-rise control signaling, physical layer signaling be arranged on synchronous code it
Afterwards；High-rise control signaling is arranged in the Deta bearer after Part I signaling.Physical layer signaling parameter include the machine colour coding,
Data type mark, time-gap number mark, slot transmission pattern identification and parity check sum flag bit, wherein data type include
Speech frame frame head, data frame frame head, data block Sub-PDU and high-rise control signaling.
The data transfer mode of the present invention can reduce overhead on the premise of system transmission performance is not reduced,
System efficiency of transmission is improved, and reduces system processing complexity, the flow of voice transfer link establishment is more rationalized.
3rd, speech data is decoded
As shown in figure 4, the detailed process of speech data decoding comprises the following steps：
Step 31, the corresponding bit stream of current speech data is obtained, and by SILK decoders to current speech data ratio
Spy's stream decodes the voice sub-band parameter after being quantified；
Step 32, extract quantify after voice sub-band parameter each sub-band surd and sonant parameter and synthesize pumping signal；The step
Rapid 32 comprise the following steps that：
Step 321, decoding end exits timer m2=0 with coding side setting simultaneously, and interval T0 is exited in setting；
Step 322, decoding end judge m2 whether the integral multiple for being T0, if it is search for initial according to index value index
Code book obtains superframe sub-band surd and sonant degree parameter；Otherwise the sub-band surd and sonant degree after being quantified using initial codebook to a upper superframe is joined
Number is quantified, and obtains index value n, and concentrates selection code book Cn from condition code book according to index value n, is received according to decoding end
Index value index search code book Cn obtains superframe sub-band surd and sonant degree parameter；
Step 323, according to superframe sub-band surd and sonant degree parameter, the sub-band surd and sonant degree parameter of each speech frame is obtained, is sent
Enter pumping signal synthesis end generation pumping signal；Make m=m+1 and go to step 322；
Step 33, the voice of synthesis is exported in real time.
Preferably, the process that the pure and impure sound decision model for judging y-bend tree construction is trained as：For pure and impure
Sound judges each problem that data are concentrated, and is to calculate voiced sound ratio value respectively with no voice training data for answering, and
And it is used as root node the problem of selection voiced sound proportional difference maximum；Enter line splitting for voice training data under root node, with
Form non-leaf nodes and leaf node.
Preferably, during the pure and impure sound decision model for judging y-bend tree construction is trained,
After division stop condition set in advance is met, stop the division, wherein the division stops bar
Part is：Non-leaf nodes or the voice training data of leaf node are less than the first threshold value set in advance；
After division stop condition set in advance is met, stop the division, wherein the division stops bar
Part is：The voiced sound proportional difference of non-leaf nodes or leaf node is less than the second threshold value set in advance.
Preferably, the pure and impure sound judges whether data set is vowel, plosive, friction as the phoneme belonging to speech frame
Sound, read again, nasal sound, high and level tone, rising tone, upper sound or falling tone, and forward and backward one in speech sentence of phoneme belonging to speech frame
Whether phoneme is vowel, plosive, fricative, stressed, nasal sound, high and level tone, rising tone, upper sound or falling tone.
Preferably, interval T0 is exited to be set according to the size of superframe.
Preferably, the speech data sampling point to acquisition is by 8kHz frequency samplings and done by high-pass filtering removal power frequency
The voice sampling point disturbed.
Preferably, the quantity y of the condition code book and the quantizing bit number x of sub-band surd and sonant degree parameter relational expression are：y
=2x。
Checked by the command methods of the present invention come the primary equipment to substation field operation, it is ensured that to correct
Primary equipment is isolated, and verification is checked using flexecutive's real-time instruction, it is ensured that the accuracy of isolation；The present invention
Convenient scene and the communication of control centre, can be sent to control centre by the sound at scene, be easy to the technical barrier to scene
Quickly solved, administrative staff's remote command site operation personnel is carried out safety operation.
The quantizing process that carried out to voice sub-band surd and sonant parameter of the present invention has taken into full account subband between successive frame (superframe)
The correlation of pure and impure tone parameter, utilizes the correlation, according to previous superframe sub-band surd and sonant degree parameter by state transfer
Quantized result, the quantization code book of adaptively selected current super frame sub-band surd and sonant degree parameter improves sub-band surd and sonant parameter
Quantitative efficiency.Meanwhile, quantizing process proposes timing escape mechanism, often crosses time T, quantifies code book and restarts source code
This, so as to effectively prevent the diffusion of channel error code, makes the influence that error code spreads be confined within time T.This method can be with
In low rate, ultralow speed speech coding algorithm effective for below 2400bps, so as to improve its synthetic speech quality.
Simply the preferred embodiment of the present invention described above, for those skilled in the art,
Without departing from the principles of the invention, some improvements and modifications can also be made, these improvements and modifications are also regarded as this hair
Bright protection domain.
Claims (7)
1. a kind of substation field operation remote command method based on voice transfer, passes through live talk back equipment and remote command
Commander's talk back equipment at center carries out interactive voice to instruct substation field operation, it is characterized in that, including procedure below：
First, encoded speech data
Step 11, obtain live talk back equipment or command the speech data of the telephone transmitter input of talk back equipment；
Step 12, the speech data of acquisition is subjected to SILK coded treatments and generates voice sub-band parameter；
Step 13, voice sub-band surd and sonant parameter is quantified；The step 13 is comprised the following steps that：
Step 131, sub-band surd and sonant in voice sub-band parameter is judged；Sub-band surd and sonant in the sub-band parameter to voice
The process judged is：Pure and impure sound is set to judge data set first；Then sentenced using voice training data and the pure and impure sound
Determine data set, the pure and impure sound decision model for judging tree construction to y-bend is trained, and wherein the y-bend judges non-in tree construction
Leaf node judges the problem of data are concentrated for the pure and impure sound, and leaf node is pure and impure sound result of determination；Finally receive voice
Test data, and judge that tone testing data, as voiceless sound or voiced sound, work as voice using the pure and impure sound decision model after training
Test data is judged as after voiced sound, and the pumping signal of the tone testing data is assumed to be impact during phonetic synthesis
Response sequence；After tone testing data are judged as voiceless sound, the excitation of the tone testing data during phonetic synthesis
Signal is assumed to be white noise；
Step 132, timer m1=0 is exited in coding side setting, and interval T0 is exited in setting；
Step 133, to the speech data sampling point framing in chronological order of acquisition, and sub-band surd and sonant degree parameter is extracted by frame；
Step 134, multiple speech frame composition superframes are combined, the sub-band surd and sonant degree parameter composition superframe subband of multiple speech frames is clear
Voiced sound degree parameter；
Step 135, judge m1 whether the integral multiple for being T0, if it is using initial codebook to superframe sub-band surd and sonant degree parameter
Carry out vector quantization；Otherwise the sub-band surd and sonant degree parameter after a upper superframe is quantified is quantified using initial codebook, it is assumed that
Quantized value is n, and the corresponding code book Cn of the index is chosen from condition codebook set according to quantization index value n, utilizes Cn pairs of code book
Current super frame sub-band surd and sonant degree parameter is quantified and obtains index value index；Described initial codebook and condition code book collection
Obtained by off-line training；
Step 136, index value is subjected to coding transmission, makes m1=m1+1 and go to step 135；
Step 14, by the corresponding bit stream of voice sub-band parameter compression generation current speech data after real-time quantization；
2nd, data transfer
Live talk back equipment or the speech data for the telephone transmitter input for commanding talk back equipment are sent to commander's talk back equipment or existing
The telephone receiver of field talk back equipment；
The detailed process of the data transfer comprises the following steps：
Step 21, telephone transmitter is when sending voice superframes, placed in the head time slot of each voice superframes synchronous code with it is complete
Signaling parameter, the signaling parameter includes all parameters for being used to carry out the foundation of voice call link；
Step 22, when telephone receiver receives signal, detect the signal that receives whether be voice superframes voice head time slot, if
It is then to be transferred to step 23, otherwise, is transferred to step 24；
Step 23, the signaling parameter of carrying is parsed from the voice head time slot, and language is carried out according to the signaling parameter parsed
The foundation of sound call link；
Step 24, according to the content for receiving signal, signal transacting is carried out；The concrete operations of the step 24 are：If received
Signal be data head time slot, then set up data interaction link；If the signal received is data block pdu, data block is solved；
If the signal received is high-rise control signaling, the high-rise control signaling of parsing；
3rd, speech data is decoded
Step 31, the corresponding bit stream of current speech data is obtained, and by SILK decoders to current speech data bit stream
Decode the voice sub-band parameter after being quantified；
Step 32, extract quantify after voice sub-band parameter each sub-band surd and sonant parameter and synthesize pumping signal；The step 32
Comprise the following steps that：
Step 321, decoding end exits timer m2=0 with coding side setting simultaneously, and interval T0 is exited in setting；
Step 322, decoding end judge m2 whether the integral multiple for being T0, if it is according to index value index search for initial codebook
Obtain superframe sub-band surd and sonant degree parameter；Otherwise the sub-band surd and sonant degree parameter after being quantified using initial codebook to a upper superframe is entered
Row quantifies, and obtains index value n, and concentrate selection code book Cn, the index received according to decoding end from condition code book according to index value n
Value index search code book Cn obtains superframe sub-band surd and sonant degree parameter；
Step 323, according to superframe sub-band surd and sonant degree parameter, the sub-band surd and sonant degree parameter of each speech frame is obtained, feeding swashs
Encourage signal synthesis end generation pumping signal；Make m=m+1 and go to step 322；
Step 33, the voice of synthesis is exported in real time.
2. a kind of substation field operation remote command method based on voice transfer according to claim 1, its feature
Be, the process that the pure and impure sound decision model for judging y-bend tree construction is trained as：Judge data set for pure and impure sound
In each problem, be to calculate voiced sound ratio value respectively with no voice training data for answering, and select voiced sound ratio
The problem of example difference is maximum is used as root node；Enter line splitting for voice training data under root node, to form non-leaf section
Point and leaf node.
3. a kind of substation field operation remote command method based on voice transfer according to claim 2, its feature
It is, during the pure and impure sound decision model for judging y-bend tree construction is trained,
After division stop condition set in advance is met, stop the division, wherein the division stop condition is：
Non-leaf nodes or the voice training data of leaf node are less than the first threshold value set in advance；
After division stop condition set in advance is met, stop the division, wherein the division stop condition is：
The voiced sound proportional difference of non-leaf nodes or leaf node is less than the second threshold value set in advance.
4. a kind of substation field operation remote command method based on voice transfer according to claim 1, its feature
Be, the pure and impure sound judge data set as the phoneme belonging to speech frame whether be vowel, plosive, fricative, read again, nasal sound,
Whether high and level tone, rising tone, upper sound or falling tone, and a forward and backward phoneme of the phoneme in speech sentence belonging to speech frame are members
Sound, plosive, fricative, stressed, nasal sound, high and level tone, rising tone, upper sound or falling tone.
5. a kind of substation field operation remote command method based on voice transfer according to claim 1, its feature
It is to exit interval T0 to be set according to the size of superframe.
6. a kind of substation field operation remote command method based on voice transfer according to claim 1, its feature
It is that the speech data sampling point to acquisition is by 8kHz frequency samplings and by the voice sampling point of high-pass filtering removal Hz noise.
7. a kind of substation field operation remote command method based on voice transfer according to claim 1, its feature
It is that the quantity y and the quantizing bit number x of sub-band surd and sonant degree parameter of the condition code book relational expression are：Y=2x。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710334346.9A CN107017007A (en) | 2017-05-12 | 2017-05-12 | A kind of substation field operation remote command method based on voice transfer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710334346.9A CN107017007A (en) | 2017-05-12 | 2017-05-12 | A kind of substation field operation remote command method based on voice transfer |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107017007A true CN107017007A (en) | 2017-08-04 |
Family
ID=59450245
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710334346.9A Pending CN107017007A (en) | 2017-05-12 | 2017-05-12 | A kind of substation field operation remote command method based on voice transfer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107017007A (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202196988U (en) * | 2011-09-16 | 2012-04-18 | 河南省电力公司周口供电公司 | Voice prompt device for protecting, observing and controlling screen cabinet of transformer substation |
| CN102855878A (en) * | 2012-09-21 | 2013-01-02 | 山东省计算中心 | Quantification method of pure and impure pitch parameters of narrow-band voice sub-band |
| CN104143342A (en) * | 2013-05-15 | 2014-11-12 | 腾讯科技（深圳）有限公司 | Voiceless sound and voiced sound judging method and device and voice synthesizing system |
| CN104469293A (en) * | 2014-11-27 | 2015-03-25 | 国网山东省电力公司泰安供电公司 | Monitoring device for remote electric operation in transformer substation |
| CN105356606A (en) * | 2015-11-21 | 2016-02-24 | 国网河南省电力公司平顶山供电公司 | Remote operation voice prompting device of transformer substation |
| CN105592077A (en) * | 2011-05-19 | 2016-05-18 | 福建联拓科技有限公司 | Voice transmission method |
| CN107256711A (en) * | 2017-05-12 | 2017-10-17 | 国网山东省电力公司经济技术研究院 | A kind of power distribution network emergency maintenance remote commanding system |
-
2017
- 2017-05-12 CN CN201710334346.9A patent/CN107017007A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105592077A (en) * | 2011-05-19 | 2016-05-18 | 福建联拓科技有限公司 | Voice transmission method |
| CN202196988U (en) * | 2011-09-16 | 2012-04-18 | 河南省电力公司周口供电公司 | Voice prompt device for protecting, observing and controlling screen cabinet of transformer substation |
| CN102855878A (en) * | 2012-09-21 | 2013-01-02 | 山东省计算中心 | Quantification method of pure and impure pitch parameters of narrow-band voice sub-band |
| CN104143342A (en) * | 2013-05-15 | 2014-11-12 | 腾讯科技（深圳）有限公司 | Voiceless sound and voiced sound judging method and device and voice synthesizing system |
| CN104469293A (en) * | 2014-11-27 | 2015-03-25 | 国网山东省电力公司泰安供电公司 | Monitoring device for remote electric operation in transformer substation |
| CN105356606A (en) * | 2015-11-21 | 2016-02-24 | 国网河南省电力公司平顶山供电公司 | Remote operation voice prompting device of transformer substation |
| CN107256711A (en) * | 2017-05-12 | 2017-10-17 | 国网山东省电力公司经济技术研究院 | A kind of power distribution network emergency maintenance remote commanding system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP3366903B2 (en) | Method and apparatus for coding an audio signal and method and apparatus for decoding a bitstream | |
| TW318239B (en) | ||
| CN108833722A (en) | Audio recognition method, device, computer equipment and storage medium | |
| CN103325375B (en) | One extremely low code check encoding and decoding speech equipment and decoding method | |
| CN105845135A (en) | Sound recognition system and method for robot system | |
| JPS603240A (en) | Voice activation detecting method | |
| CN112365882A (en) | Speech synthesis method, model training method, device, equipment and storage medium | |
| CN105206271A (en) | Intelligent equipment voice wake-up method and system for realizing method | |
| US8571204B2 (en) | Apparatus and method for echo control in parameter domain | |
| CN103533129B (en) | Real-time voiced translation communication means, system and the communication apparatus being applicable | |
| CN101273403A (en) | Scalable encoding apparatus, scalable decoding apparatus, and methods of them | |
| WO2020238058A1 (en) | Voice transmission method and apparatus, computer device and storage medium | |
| CN105895082A (en) | Acoustic model training method and device as well as speech recognition method and device | |
| CN101719966A (en) | Monitoring method of call tone quality, monitoring terminal and system therefor | |
| CN110444224A (en) | A kind of method of speech processing and device based on production confrontation network | |
| CN107256711A (en) | A kind of power distribution network emergency maintenance remote commanding system | |
| CN107017007A (en) | A kind of substation field operation remote command method based on voice transfer | |
| CN117713377A (en) | Intelligent voice joint debugging system of dispatching automation master station | |
| CN107464569A (en) | Vocoder | |
| CN103854655A (en) | Low-bit-rate voice coder and decoder | |
| Yong | Study of voice packet reconstruction methods applied to CELP speech coding | |
| CN110533750A (en) | A method of it converts the audio into as the sign language animation with customized 3D role | |
| CN106325260A (en) | Rolling mill transmission trip fault voice alarm system | |
| KR100984094B1 (en) | A voiced/unvoiced decision method for the smv of 3gpp2 using gaussian mixture model | |
| Nabil et al. | Formant distortion after codecs for Arabic |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170804 |