CN104766605B - Clock synchronization system and method between a kind of locomotive recording device and LKJ supervising devices - Google Patents
Clock synchronization system and method between a kind of locomotive recording device and LKJ supervising devices Download PDFInfo
- Publication number
- CN104766605B CN104766605B CN201510194350.0A CN201510194350A CN104766605B CN 104766605 B CN104766605 B CN 104766605B CN 201510194350 A CN201510194350 A CN 201510194350A CN 104766605 B CN104766605 B CN 104766605B
- Authority
- CN
- China
- Prior art keywords
- voice message
- time
- voice
- message signal
- signal
- 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.)
- Active
Links
Landscapes
- Electric Clocks (AREA)
Abstract
The present invention relates to the clock synchronization system between a kind of locomotive recording device and LKJ supervising devices and method, present invention firstly receives the voice signal of locomotive recording device input, and therefrom gathers the voice message signal that LKJ supervising devices are sent；Then the voice signal to input and voice message signal carry out end-point detection and feature extraction；The feature of voice message signal is trained, the feature of input speech signal is matched and identified with training pattern, at the time of identifying that voice message signal in input speech signal and the voice message signal are in sound recordings；Synchronized at the time of finally by the moment where the voice message signal in the sound recordings detected with where the voice message signal in LKJ log sheets, so as to realize the time synchronized of locomotive recording device and LKJ.The present invention realizes the synchronization between two devices, time and difficulty when greatly reducing school on the basis of not increasing the cost of recording device by way of Speech processing.
Description
Technical field
The present invention relates to the clock synchronization system between a kind of locomotive recording device and LKJ supervising devices and method, belong to language
Sound signal processing technology field.
Background technology
In order to strengthen trainman's standardized work consciousness, it is ensured that motorcycle safety travels, and each trainman can be equipped with
There is locomotive recording device, whole process records calling respond, station-locomotive joint control and other voice handling situations in locomotive running.
LKJ is train operation monitoring and recording device, is domestic train overspeeding protection's equipment, it can record exactly train operating conditions,
Signalling arrangement situation and crew manipulate situation, and relatively reliable using two-node cluster hot backup redundancy of effort mode, service behaviour；Device
Screen display the information such as forward box situation, running situation are shown in a manner of figure, curve, word etc..
And LKJ is while be used for train speed and control, the acquisition and recording various kinds of vehicles relevant with safe train operation is run
Status information, and provide related voice prompting.Locomotive work personnel can be according to the recorded message and LKJ logs of recording device
File is operated against regulations to judge trainman whether there is.But in a practical situation, the time of locomotive recording device and LKJ time
It is inconsistent, cause recorded message time and LKJ log sheet Time Inconsistencies, and record length is usually several small
When, locomotive work personnel, by being adjusted by hand, especially lose time to the time of recording device.
And in the prior art to realize synchronization between two devices, or during using artificial school, or by means of GPS.
For when artificial school, substantial amounts of manpower is not only needed, and efficiency comparison is low, it is also possible to because human factor causes pair
When there is error；For the mode by means of GPS, although the synchronization between two devices can be realized quickly and easily,
The cost of recording device can be synchronously greatly increased by way of increasing hardware.
The content of the invention
It is an object of the invention to provide the clock synchronization system between a kind of locomotive recording device and LKJ supervising devices and side
Method, during solving to use artificial school between current locomotive recording device and LKJ supervising devices the time grow, the problem of efficiency is low.
The present invention additionally provides the time between a kind of locomotive recording device and LKJ supervising devices for solution above-mentioned technical problem
Synchronization system, the clock synchronization system include voice input module, voice message signal acquisition module, Speech processing mould
Block, sound identification module and time synchronized module,
One output end of the voice input module is directly connected with the input of voice signal processing module, and voice is defeated
Another output end for entering module is connected by the input of voice message signal acquisition module and voice signal processing module, is used
Voice message signal in the voice signal in voice input module and voice message signal acquisition module carries out spy respectively
Sign extraction, the output end of voice signal processing module and the input of sound identification module connect, for the feature according to extraction
At the time of voice message signal and the voice message signal included in identified input voice signal are in sound recordings, voice
The output end of identification module is connected with the input of time synchronized module, and the voice for being identified according to sound identification module carries
Time synchronized is carried out to locomotive recording device and LKJ supervising devices at the time of showing signal.
The synchronization module be the voice message signal that will identify that with it is all right in LKJ supervising device log sheets
It is compared at the time of answering voice signal place, at the time of determining time difference minimum, using the moment as in sound recordings device
At the time of where the voice message signal selected, and the time in recording device is corrected with the identified time difference, realize record
Time synchronized between mixer and LKJ supervising devices.
Described voice signal processing module includes end-point detection submodule and feature extraction submodule, the end-point detection
Submodule is used for the starting point and terminal for detecting voice signal and voice message signal, and the feature extraction submodule is used to extract
The feature of voice signal and voice message signal after the detection of end-point detection submodule.
The voice message signal that the LKJ supervising devices of the voice message signal acquisition module collection are sent can be one
Individual or multiple different voice message signal or multiple same voice cues at different moments.
For the voice message signal of voice message signal acquisition module collection, if a voice message signal, institute
State time synchronized module and find out corresponding and moment immediate voice message signal in LKJ supervising device log sheets,
If multiple voice message signals, the time synchronized module is by calculating time at above-mentioned multiple voice message signal moment
Difference, found out in LKJ supervising device log sheets it is closest with the voice message signal moment respectively, and the time difference with it is upper
The time difference of calculating equal multiple voice message signals are stated, at the time of by voice message signal in LKJ log sheets
At the time of replacing corresponding voice message signal in sound recordings respectively, pass through sound recordings and LKJ supervising device logs
File corresponds to the time difference of voice message signal, corrects the time in recording device so that the time of recording device and LKJ monitoring
The time synchronized of device.
Present invention also offers the method for synchronizing time between a kind of locomotive recording device and LKJ supervising devices, the synchronization side
Method comprises the following steps：
1) voice signal of locomotive recording device input is received, and gathers the voice message that wherein LKJ supervising devices are sent
Signal；
2) feature extraction is carried out respectively to the voice signal in locomotive recording device and voice message signal；
3) by the feature of the voice signal in the locomotive recording device extracted and the progress of the feature of voice message signal
Match somebody with somebody, identify voice message signal in input speech signal and the voice message signal in locomotive recording device sound recordings
At the time of；
4) finally by moment where the voice message signal in the sound recordings detected and LKJ supervising device logs
Synchronized at the time of voice message signal place is corresponded in file, to realize the time synchronized of locomotive recording device and LKJ.
In the step 4) is synchronously in the voice message signal and LKJ supervising device log sheets that will identify that
It is compared at the time of where all corresponding voice signals, at the time of determining time difference minimum, using the moment as sound recordings
At the time of where the voice indication signal selected in device, and the time in recording device is corrected with the identified time difference,
Realize the time synchronized between locomotive recording device and LKJ supervising devices.
Characteristic extraction procedure in the step 2) includes end-point detection subprocess and feature extraction subprocess, the end points
Detection subprocess is used for the starting point and terminal for detecting voice signal and voice message signal, and the feature extraction subprocess is used for
Extract the feature of voice signal and voice message signal after the detection of end-point detection submodule.
The voice message signal that collection LKJ supervising devices are sent in the step 1) can be one, or it is multiple not
Same voice message signal or multiple same voice cues at different moments.
When voice message signal for collection in step 1) is two or more, the step 4) is by calculating above-mentioned two
The time difference at individual or multiple voice message signal moment, find out in LKJ supervising device log sheets and believe with voice message
Number moment is closest, and two or more voice message signals that the time difference is equal with the time difference of above-mentioned calculating, and LKJ is transported
At the time of replacing corresponding voice message signal in sound recordings respectively at the time of voice message signal in row log file, lead to
Cross sound recordings and time difference at the time of LKJ supervising device log sheets correspond to voice message signal, amendment recording dress
Time in putting so that the time of recording device and the time synchronized of LKJ supervising devices.
The beneficial effects of the invention are as follows:The present invention gathers from the voice signal for receiving the input of locomotive recording device first
The voice message signal that LKJ supervising devices are sent；Then the voice signal to input and voice message signal carry out end-point detection
And feature extraction；The feature of voice message signal is trained, the feature of input speech signal is matched with training pattern
And identification, identify the voice message signal in input speech signal and the voice message signal at the time of during recorder is recorded；
Finally the moment where the voice message signal in the sound recordings detected is believed with the voice message in LKJ log sheets
Synchronized at the time of where number, so as to realize the time synchronized of locomotive recording device and LKJ.The present invention by voice signal at
The mode of reason realizes the synchronization between two devices, when greatly reducing school on the basis of not increasing the cost of recording device when
Between and difficulty.
Brief description of the drawings
Fig. 1 is the structural representation of the clock synchronization system between locomotive recording device and LKJ supervising devices of the present invention；
Fig. 2 is the flow chart of the method for synchronizing time between locomotive recording device and LKJ supervising devices of the present invention.
Embodiment
The embodiment of the present invention is further described below in conjunction with the accompanying drawings.
The embodiment of clock synchronization system between a kind of the locomotive recording device and LKJ supervising devices of the present invention
The synchronization system of the present invention is used to synchronize the time between locomotive recording device and LKJ supervising devices, specifically
Structure is as shown in figure 1, including voice input module, voice message signal acquisition module, voice signal processing module, speech recognition
Module and time synchronized module, an output end of voice input module directly connect with the input of voice signal processing module
Connect, the input that another output end of voice input module passes through voice message signal acquisition module and voice signal processing module
End connection, for entering to the voice message signal in the voice signal in voice input module and voice message signal acquisition module
Row end-point detection and feature extraction, the output end of speech processing module and the input of sound identification module connect, for identifying
At the time of voice message signal and the voice message signal included in input speech signal are in sound recordings, speech recognition
The output end of module is connected with the input of time synchronized module, for the voice message letter identified according to sound identification module
Number at the time of time synchronized is carried out to locomotive recording device and LKJ supervising devices.Wherein the specific implementation process of modules is such as
Under：
Voice input module is used for the voice signal for receiving the input of locomotive recording device, and voice signal here includes locomotive
Calling respond, station-locomotive joint control and other voice handling situations in running, also include the language that LKJ supervising devices are sent
Sound cue, the locomotive recording device in the present embodiment can use recording pen.
Voice message signal acquisition module is used to gather the language that LKJ is sent in the voice signal that voice input module is received
Sound cue.Voice message signal in the present embodiment is LKJ " green light passes through " voice signals sent or " red
Amber light ", " double amber lights " etc., as long as the voice message that LKJ is sent, and the voice signal can that recording device is recorded, this
In voice message signal is used as using " green light by " voice signal that LKJ supervising devices are sent.
Voice signal processing module includes end-point detection submodule and feature extraction submodule, wherein end-point detection submodule
For detecting the starting point and terminal of voice signal and voice message signal, short-time energy, short-time average zero-crossing rate can be used, fallen
The modes such as spectrum signature, entropy, complexity (KC complexity and C0 complexity) are realized.This implementation uses short-time energy and short-time average mistake
The detection method of zero rate, short-time energy and short-time average zero-crossing rate is also referred to as double-threshold comparing method, and it is detected in short-time energy
On the basis of method, plus short-time average zero-crossing rate, detected using energy and zero-crossing rate as feature, in signal to noise ratio not
In the case of being very low, the hypothesis of noise energy is more than according to the energy of voice signal, by the energy of comparator input signal with
The size of speech energy threshold value, voice segments and non-speech segment can be distinguish between.Feature extraction submodule is above-mentioned for extracting
The feature of voice signal and voice message signal after end-point detection, MFCC (mel cepstrum coefficients), LPCC can be used (linear pre-
Survey cepstrum coefficient) realize, that selected in this implementation is MFCC, and the skill of comparative maturity is belonged to using MFCC extractions voice messaging feature
Art, no longer describe in detail here.
The feature that sound identification module is used for the voice message signal obtained to signal processing module carries out machine learning instruction
Practice, the feature of the feature of above-mentioned input speech signal and voice message signal is matched and identified, so as to identified input language
Voice message signal included in sound signal, and at the time of determine that the voice message signal is in sound recordings.
At the time of time synchronized module is used for the voice message signal identified according to sound identification module, run with LKJ
Synchronized at the time of in log file where corresponding voice signal.Voice in sound recordings is detected in the present embodiment first
It is compared at the time of at the time of where cue with where corresponding voice message signal in LKJ log sheets, so
Afterwards by calculate find out corresponding voice message signal time in sound recordings and LKJ log sheets it is poor minimum at the time of,
Finally make locomotive recording device and LKJ time synchronizeds using the minimum time found the poor and corresponding moment.
The voice message signal that voice message signal acquisition module is gathered can also be two that LKJ supervising devices are sent
The different voice message of the above, now time synchronized module can be by calculating the two (or multiple) voice message signal moment
Time difference, found out in LKJ log sheets it is closest with the two (or multiple) voice message signal moment, and
Time difference two (or multiple) voice message signals equal with the time difference of above-mentioned calculating, by LKJ log sheets
Voice message signal at the time of at the time of replace corresponding voice message signal in sound recordings respectively, while pass through note of recording
Record corresponds to the time difference at two moment of voice message signal with LKJ log sheets, corrects the time in recording device, makes
Obtain the time of recording device and the time synchronized of LKJ supervising devices.
The embodiment of method for synchronizing time between a kind of the locomotive recording device and LKJ supervising devices of the present invention
As shown in Fig. 2 the method for synchronizing time in the present embodiment is that the locomotive equipment time is entered by speech recognition technology
Row simultaneously match, the voice that collection LKJ supervising devices are sent first from the voice signal for receiving the input of locomotive recording device carry
Show signal；Then the voice signal to input and voice message signal carry out end-point detection and feature extraction；Voice message is believed
Number feature be trained, the feature of input speech signal is matched and identified with training pattern, identify input voice letter
At the time of voice message signal and the voice message signal in number are in sound recordings；Finally by the sound recordings detected
Voice message signal where the moment with LKJ log sheets the voice message signal place at the time of synchronize, from
And realize locomotive recording device and LKJ time synchronized.The detailed process of this method is as follows：
1. receiving the voice signal of locomotive recording device input, the locomotive recording device in the present embodiment can use recording
Pen, calling respond, station-locomotive joint control and other voice handling situations in locomotive running are recorded by recording pen whole process, together
When, recording pen also records the voice message signal that LKJ supervising devices are sent.
2. the voice message signal that LKJ supervising devices are sent is gathered from the voice signal of input.Language in the present embodiment
Sound cue is " green light passes through " voice signal or " reddish yellow lamp ", " double amber lights " etc. that LKJ is sent, as long as LKJ
The voice message sent, and the voice signal can that recording device is recorded.This implementation is sent " green with LKJ supervising devices
Lamp by " voice signal is used as voice message signal.
3. the voice signal and voice message signal of pair input carry out end-point detection and feature extraction.
End-point detection can use short-time energy, short-time average zero-crossing rate, cepstrum feature, entropy, complexity (KC complexity and C0
Complexity) etc. mode realize.Short-time energy and short-time average zero-crossing rate, short-time energy and short-time average zero-crossing rate are selected in this implementation
Detection method be also referred to as double-threshold comparing method, it is on the basis of short-time energy detection method, plus short-time average zero passage
Rate, detected using energy and zero-crossing rate as feature, in the case where signal to noise ratio is not very low, according to voice signal
Energy is more than the hypothesis of noise energy, can be to voice by the energy and the size of speech energy threshold value of comparator input signal
Section and non-speech segment are distinguish between.Feature extraction can use MFCC (mel cepstrum coefficients), LPCC (linear prediction residue error)
Realize, that selected in this implementation is MFCC, the technology of comparative maturity is belonged to using MFCC extractions voice messaging feature, here no longer
Describe in detail.
By detecting rising for each voice segments in " green light passes through " voice signal and input speech signal in the present embodiment
Initial point and terminal, extract the phonetic feature of voice segments in " green light passes through " voice signal and input signal after end-point detection.
4. the feature of pair voice message signal carries out machine learning training, feature and the training pattern of input speech signal are entered
Row matching and identification, identify the voice message signal in input speech signal, and determine that the voice message signal is remembered in recording
At the time of in record.Machine learning training, input voice letter are carried out to " green light passes through " phonic signal character of extraction in this implementation
Number feature matched and identified with training pattern, identify in input speech signal own " green light passes through " voice letter
Number, and at the time of determine where each " green light passes through " voice signal.
5. according to voice identification result, the moment where the voice message signal in the sound recordings detected is transported with LKJ
Synchronized at the time of in row log file where the voice message signal.
Transported at the time of the present embodiment is arbitrarily selected in input speech signal where " green light a passes through " voice signal with LKJ
It is compared at the time of owning " green light passes through " voice signal place in row log file, will at the time of finding out time difference minimum
At the time of the moment is as where " green light passes through " voice signal selected in sound recordings, while pass through sound recordings and LKJ
Log sheet corresponds to the time difference at two moment of voice signal, corrects the time in recording device, make recording device and
The time synchronized of LKJ supervising devices.
The voice message signal gathered can also be the different voice message of the two or more that sends of LKJ supervising devices,
It can now be looked for by calculating time difference at the two (or multiple) voice message signal moment in LKJ log sheets
Go out closest with the two (or multiple) voice message signal moment, and the time difference is equal with the time difference of above-mentioned calculating
Two (or multiple) voice message signals, record is replaced respectively at the time of by voice message signal in LKJ log sheets
In sound record at the time of corresponding voice message signal, while voice is corresponded to by sound recordings and LKJ log sheets and carried
Show the time difference at two moment of signal, correct the time in recording device so that the time of recording device and LKJ supervising devices
Time synchronized.
Method during compared to existing craft to locomotive recording device and LKJ supervising device schools, utilize the machine of the present invention
Clock synchronization system and method between car recording device and LKJ supervising devices, only need to be realized by software locomotive recording device and
The automatic synchronization of LKJ times, time and difficulty when greatly reducing school on the basis of not increasing the cost of recording device.
Claims (10)
- A kind of 1. clock synchronization system between locomotive recording device and LKJ supervising devices, it is characterised in that the clock synchronization system Including voice input module, voice message signal acquisition module, voice signal processing module, sound identification module and time synchronized Module,One output end of the voice input module is directly connected with the input of voice signal processing module, phonetic entry mould Another output end of block is connected by voice message signal acquisition module with the input of another voice signal processing module, For being carried out respectively to the voice message signal in the voice signal in voice input module and voice message signal acquisition module Feature extraction, the output end of voice signal processing module and the input of sound identification module connect, for the spy according to extraction Levy identified input voice signal included in voice message signal and the voice message signal in sound recordings at the time of, language The output end of sound identification module is connected with the input of time synchronized module, for the voice identified according to sound identification module Time synchronized is carried out to locomotive recording device and LKJ supervising devices at the time of cue.
- 2. the clock synchronization system between locomotive recording device according to claim 1 and LKJ supervising devices, its feature exist In the synchronization module is the voice message signal and all corresponding languages in LKJ supervising device log sheets that will identify that It is compared at the time of where sound signal, at the time of determining time difference minimum, using the moment as selected in sound recordings device At the time of where fixed voice message signal, and the time in recording device is corrected with the identified time difference, realize recording dress Put the time synchronized between LKJ supervising devices.
- 3. the clock synchronization system between locomotive recording device according to claim 2 and LKJ supervising devices, its feature exist In described voice signal processing module includes end-point detection submodule and feature extraction submodule, the end-point detection submodule Block is used for the starting point and terminal for detecting voice signal and voice message signal, and the feature extraction submodule is used to extract through end The feature of voice signal and voice message signal after point detection sub-module detection.
- 4. the clock synchronization system between locomotive recording device according to claim 3 and LKJ supervising devices, its feature exist In, the voice message signal that the LKJ supervising devices of the voice message signal acquisition module collection are sent can be one, or Multiple different voice message signals or multiple same voice cues at different moments.
- 5. the clock synchronization system between locomotive recording device according to claim 4 and LKJ supervising devices, its feature exist In for the voice message signal of voice message signal acquisition module collection, if a voice message signal, the time Synchronization module finds out corresponding and moment immediate voice message signal in LKJ supervising device log sheets, if Multiple voice message signals, the time synchronized module by calculating time difference at above-mentioned multiple voice message signal moment, Closest with the voice message signal moment respectively, and time difference and above-mentioned meter are found out in LKJ supervising device log sheets Time difference of calculation equal multiple voice message signals, distinguish at the time of by voice message signal in LKJ log sheets At the time of replacing corresponding voice message signal in sound recordings, pass through sound recordings and LKJ supervising device log sheets The time difference of corresponding voice message signal, correct the time in recording device so that the time of recording device and LKJ supervising devices Time synchronized.
- 6. the method for synchronizing time between a kind of locomotive recording device and LKJ supervising devices, it is characterised in that the synchronous method includes Following steps：1) voice signal of locomotive recording device input is received, and gathers the voice message signal that wherein LKJ supervising devices are sent；2) feature extraction is carried out respectively to the voice signal in locomotive recording device and voice message signal；3) feature of the voice signal in the locomotive recording device extracted is matched with the feature of voice message signal, known Do not go out the voice message signal in input speech signal and the voice message signal in locomotive recording device sound recordings when Carve；4) finally by moment where the voice message signal in the sound recordings detected and LKJ supervising device log sheets Synchronized at the time of where middle corresponding voice message signal, to realize the time synchronized of locomotive recording device and LKJ.
- 7. the method for synchronizing time between locomotive recording device according to claim 6 and LKJ supervising devices, its feature exist In in the step 4) is synchronously the voice message signal that will identify that with owning in LKJ supervising device log sheets It is compared at the time of where corresponding voice signal, at the time of determining time difference minimum, using the moment as sound recordings device At the time of where middle selected voice indication signal, and the time in recording device is corrected with the identified time difference, realized Time synchronized between locomotive recording device and LKJ supervising devices.
- 8. the method for synchronizing time between locomotive recording device according to claim 7 and LKJ supervising devices, its feature exist In the characteristic extraction procedure in the step 2) includes end-point detection subprocess and feature extraction subprocess, the end-point detection Subprocess is used for the starting point and terminal for detecting voice signal and voice message signal, and the feature extraction subprocess is used to extract The feature of voice signal and voice message signal after the detection of end-point detection submodule.
- 9. the method for synchronizing time between locomotive recording device according to claim 8 and LKJ supervising devices, its feature exist In the voice message signal that collection LKJ supervising devices are sent in the step 1) can be one, or multiple different languages Sound cue or multiple same voice cues at different moments.
- 10. the method for synchronizing time between locomotive recording device according to claim 9 and LKJ supervising devices, its feature exist In, for collection in step 1) voice message signal for it is two or more when, the step 4) by calculate above-mentioned two or The time difference at multiple voice message signal moment, when being found out in LKJ supervising device log sheets with voice message signal Closest, and two or more voice message signals that the time difference is equal with the time difference of above-mentioned calculating are carved, LKJ is run and remembered At the time of replacing corresponding voice message signal in sound recordings respectively at the time of recording the voice message signal in file, pass through record Sound record corresponds to time difference at the time of voice message signal with LKJ supervising device log sheets, corrects in recording device Time so that the time of recording device and the time synchronized of LKJ supervising devices.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510194350.0A CN104766605B (en) | 2015-04-23 | 2015-04-23 | Clock synchronization system and method between a kind of locomotive recording device and LKJ supervising devices |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510194350.0A CN104766605B (en) | 2015-04-23 | 2015-04-23 | Clock synchronization system and method between a kind of locomotive recording device and LKJ supervising devices |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104766605A CN104766605A (en) | 2015-07-08 |
| CN104766605B true CN104766605B (en) | 2018-03-27 |
Family
ID=53648389
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510194350.0A Active CN104766605B (en) | 2015-04-23 | 2015-04-23 | Clock synchronization system and method between a kind of locomotive recording device and LKJ supervising devices |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104766605B (en) |
Families Citing this family (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105262976B (en) * | 2015-09-22 | 2019-02-01 | 武汉钢铁(集团)公司 | A kind of control third party's recorded broadcast source method and multipoint controller MCU |
| CN106813702B (en) * | 2015-12-02 | 2019-08-13 | 北京铁路信号有限公司 | A kind of test device and system of communications records device |
| CN105741855B (en) * | 2016-02-03 | 2019-12-20 | 郑州畅想高科股份有限公司 | Attendant call response detection method based on audio analysis |
| CN105843884B (en) * | 2016-03-21 | 2019-05-24 | 西安银石科技发展有限责任公司 | One kind being based on LKJ and its audio-visual synchronization intelligent retrieval analysis method |
| CN106657852B (en) * | 2016-12-21 | 2019-04-16 | 智威富（北京）科技有限公司 | Time synchronization method, equipment and system |
| CN106601241A (en) * | 2016-12-26 | 2017-04-26 | 河南思维信息技术有限公司 | Automatic time correcting method for recording file |
| CN106649745A (en) * | 2016-12-26 | 2017-05-10 | 河南思维信息技术有限公司 | Locomotive application comprehensive analysis method |
| CN109545027B (en) * | 2018-12-24 | 2021-06-01 | 郑州畅想高科股份有限公司 | Training platform, crew simulation training method and device |
| CN113763988B (en) * | 2020-06-01 | 2024-05-28 | 中车株洲电力机车研究所有限公司 | Time synchronization method and system for locomotive cab monitoring information and LKJ monitoring information |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007047291A (en) * | 2005-08-08 | 2007-02-22 | Matsushita Electric Ind Co Ltd | Voice interpretation system and voice interpretation method |
| CN200959133Y (en) * | 2006-06-23 | 2007-10-10 | 北京华韬致诚科技有限责任公司 | Apparatus for phonetic record and telecommunication of locomotive |
| CN101186216A (en) * | 2006-11-17 | 2008-05-28 | 兰州交通大学 | Switching locomotive operation safety monitoring apparatus |
| CN104123932A (en) * | 2014-07-29 | 2014-10-29 | 科大讯飞股份有限公司 | Voice conversion system and method |
-
2015
- 2015-04-23 CN CN201510194350.0A patent/CN104766605B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007047291A (en) * | 2005-08-08 | 2007-02-22 | Matsushita Electric Ind Co Ltd | Voice interpretation system and voice interpretation method |
| CN200959133Y (en) * | 2006-06-23 | 2007-10-10 | 北京华韬致诚科技有限责任公司 | Apparatus for phonetic record and telecommunication of locomotive |
| CN101186216A (en) * | 2006-11-17 | 2008-05-28 | 兰州交通大学 | Switching locomotive operation safety monitoring apparatus |
| CN104123932A (en) * | 2014-07-29 | 2014-10-29 | 科大讯飞股份有限公司 | Voice conversion system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104766605A (en) | 2015-07-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104766605B (en) | Clock synchronization system and method between a kind of locomotive recording device and LKJ supervising devices | |
| CN110867196A (en) | Machine equipment state monitoring system based on deep learning and voice recognition | |
| CN105424395B (en) | Method and device for determining equipment fault | |
| CN110940539A (en) | Machine equipment fault diagnosis method based on artificial experience and voice recognition | |
| CN108428448A (en) | A kind of sound end detecting method and audio recognition method | |
| CN106250822A (en) | Student's focus based on recognition of face monitoring system and method | |
| CN107305541A (en) | Speech recognition text segmentation method and device | |
| CN102682760A (en) | Overlapped voice detection method and system | |
| CN108776452B (en) | Special equipment field maintenance monitoring method and system | |
| CN103220050A (en) | Device and method for CIR comprehensive diagnosis test | |
| CN105741855B (en) | Attendant call response detection method based on audio analysis | |
| CN112784695A (en) | Driver abnormal state detection method based on image and voice recognition | |
| CN106098079A (en) | Method and device for extracting audio signal | |
| CN105575402A (en) | Network teaching real time voice analysis method | |
| CN109545027A (en) | A kind of practical traning platform, crew's simulation training method and device | |
| CN106328165A (en) | Robot autologous sound source elimination system | |
| CN112801466B (en) | Method and system for early warning illegal operation of oil discharge operation of gas station | |
| CN107569246B (en) | Embedded controller fatigue risk measuring device and method thereof | |
| CN108419200A (en) | A kind of microphone array fault detection method and device | |
| CN108389281A (en) | A kind of unmanned plane cruising inspection system with voice record function | |
| ATE332065T1 (en) | DISTRIBUTED TEST SYSTEM FOR TESTING AND MONITORING ELECTRONIC SIGNALS ON NETWORK CONNECTIONS | |
| CN113763988B (en) | Time synchronization method and system for locomotive cab monitoring information and LKJ monitoring information | |
| CN104680275A (en) | Management method for public security examination based on internet technology | |
| AU2020385426A1 (en) | Multi-layer coupling relationship-based method for identifying train operation deviation propagation conditions | |
| Lashkari et al. | NMF-based cepstral features for speech emotion recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |