[go: up one dir, main page]
More Web Proxy on the site http://driver.im/

CN106297771A - Multifunctional calling extension set - Google Patents

Multifunctional calling extension set Download PDF

Info

Publication number
CN106297771A
CN106297771A CN201610670271.7A CN201610670271A CN106297771A CN 106297771 A CN106297771 A CN 106297771A CN 201610670271 A CN201610670271 A CN 201610670271A CN 106297771 A CN106297771 A CN 106297771A
Authority
CN
China
Prior art keywords
unit
layer
speech
main control
pretreatment
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
Application number
CN201610670271.7A
Other languages
Chinese (zh)
Other versions
CN106297771B (en
Inventor
曹善江
程志强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Huasheng Medical Technology Co.,Ltd.
Original Assignee
Yantai Beacon Medical Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Yantai Beacon Medical Technology Co Ltd filed Critical Yantai Beacon Medical Technology Co Ltd
Priority to CN201610670271.7A priority Critical patent/CN106297771B/en
Publication of CN106297771A publication Critical patent/CN106297771A/en
Application granted granted Critical
Publication of CN106297771B publication Critical patent/CN106297771B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M11/00—Telephonic communication systems specially adapted for combination with other electrical systems
    • H04M11/02—Telephonic communication systems specially adapted for combination with other electrical systems with bell or annunciator systems
    • H04M11/022—Paging systems
    • 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
    • G10L15/02—Feature extraction for speech recognition; Selection of recognition unit
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/08—Speech classification or search
    • G10L15/16—Speech classification or search using artificial neural networks
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • G10L21/0208—Noise filtering
    • G10L21/0216—Noise filtering characterised by the method used for estimating noise
    • G10L21/0232—Processing in the frequency domain
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M11/00—Telephonic communication systems specially adapted for combination with other electrical systems
    • H04M11/02—Telephonic communication systems specially adapted for combination with other electrical systems with bell or annunciator systems
    • H04M11/027—Annunciator systems for hospitals

Landscapes

  • Engineering & Computer Science (AREA)
  • Acoustics & Sound (AREA)
  • Computational Linguistics (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Human Computer Interaction (AREA)
  • Physics & Mathematics (AREA)
  • Signal Processing (AREA)
  • Multimedia (AREA)
  • Artificial Intelligence (AREA)
  • Evolutionary Computation (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Rehabilitation Tools (AREA)

Abstract

The invention provides a kind of Multifunctional calling extension set, including main control chip, also include mike, speaker and the receiver module being each connected respectively with main control chip, described main control chip connects also by level cache discrimination module, and described discrimination module is connected by L2 cache sending module；Described discrimination module, based on fpga chip, builds the BP neutral net for differentiating invalid voice data slot in described fpga chip.The present invention determines invalid voice data slot according to the characteristic information of speech data fragment, transmitted to main frame by sending module by L2 cache again after then invalid voice data slot is replaced with the blank sound data that volume is minimum, have compressed the size of transmitting audio data, reduce the calling extension set demand to bandwidth.

Description

Multifunctional calling extension set
Technical field
The present invention relates to Multifunctional calling extension set, belong to field of medical device.
Background technology
Calling extension set mainly coordinates nurse station main frame to use, and intercommunication communication between main frame.
Owing to the berth of hospital is more, the network bandwidth cannot meet the requirement of many extension sets simultaneous communications.Prior art does not has It is provided with a solution and carries out screening, compression to reduce the demand to bandwidth for calling extension set being transmitted data.
Summary of the invention
For reducing the calling extension set demand to bandwidth, the present invention proposes a kind of Multifunctional calling extension set.
Technical solution of the present invention is as follows:
Multifunctional calling extension set, including main control chip, also includes the mike being each connected with main control chip respectively, raises Sound device and receiver module, described main control chip connects also by level cache discrimination module, and described discrimination module passes through two Level caching connection has sending module；
Described discrimination module, based on fpga chip, has been built in described fpga chip for differentiating invalid voice data slot BP neutral net, described BP neutral net differentiate invalid voice data slot method step be:
(A) voice that mike is included by main control chip is converted into speech data, by below 50Hz in this speech data and The frequency range of more than 1200Hz all filters, then the speech data after filtering is divided into speech-sound data sequence will in units of 3s This speech-sound data sequence leaves in level cache；
Main control chip is successively read the element in speech-sound data sequence from level cache, is handled as follows respectively:
(A-1) the ensemble average decibel value remembering this element is x1, overall code check be x2；
(A-2) this element is carried out frequency-domain analysis, with 50Hz as starting point, calculate the rate of change often crossing 50Hz decibel value, note Recording first rate of change frequency values more than 0.1dB/Hz is x3, first rate of change frequency values less than-0.1dB/Hz be x4； If not finding qualified x3, then by x3It is set as 50Hz, if not finding qualified x4, then by x4It is set as 1200Hz；
(A-3) x is calculated3To x4The average decibel value of frequency range is x5；
(A-4) by x1、x2、x3、x4And x5Store to level cache as one group of input data；
(B) it is sequentially delivered in the BP neutral net of discrimination module sentence by each group input data in level cache Not；
This BP neutral net is disposed with input layer, pretreatment layer, intermediate layer and output layer along input to outbound course；
Described input layer includes for inputting x1Input block one, be used for inputting x2Input block two, be used for inputting x3 Input block three, be used for inputting x4Input block four and be used for inputting x5Input block five；
Described pretreatment layer includes pretreatment unit one, pretreatment unit two, pretreatment unit three and pretreatment unit four；
Described intermediate layer includes temporary location one, temporary location two and temporary location three；
Described output layer includes output unit；
Described input layer, pretreatment layer, intermediate layer and output layer are respectively the 1st layer of BP neutral net, the 2nd layer, the 3rd layer With the 4th layer；
Described input block one, input block two, input block three, input block four and input block five are respectively the 1st Unit the 1st, Unit the 2nd, Unit the 3rd, Unit the 4th and Unit the 5th of layer；
Described pretreatment unit one, pretreatment unit two, pretreatment unit three and pretreatment unit four are respectively the 2nd layer Unit the 1st, Unit the 2nd, Unit the 3rd and Unit the 4th；
Described temporary location one, temporary location two and temporary location three are respectively Unit the 1st, Unit the 2nd and the of the 3rd layer Unit 3；
Described output unit is Unit the 1st of the 4th layer；
If the output valve of l layer i-th cell isBias term isActivation primitive is fi (l)(), the unit of l layer Sum is n(l), the output valve of l layer jth unitWeights when being transferred to l+1 layer i-th cell are
Then for the 1st layer:
For the 2nd to 4 layer:
IfWithPerseverance is 0；
According to the data of input, BP neutral net judges whether this element is invalid voice data slot, if invalid language This element is then replaced with blank sound data by sound data slot；
(C) speech-sound data sequence that replacement was processed by discrimination module is sent in L2 cache.
Further: the activation primitive of described pretreatment layer each unit is:
f i ( l ) ( x ) = e 3 x + 3 x 3 - 1 e 2 x + 1 - x , x &GreaterEqual; 0 sin ( x 5 ) , x < 0 .
Further: the activation primitive of described intermediate layer and output layer each unit is:
fi (l)(x)=max (0, x-0.2).
Further, the training method of BP neutral net is: background noise frequency be 20Hz, 80Hz, 150Hz, 200Hz, 360Hz, 500Hz, 750Hz, 1000Hz, 1500Hz, 2000Hz and each all without recording 100 in the environment of voice respectively The invalid sample voice data of bar duration 30s, and background noise frequency be 20Hz, 80Hz, 150Hz, 200Hz, 360Hz, 500Hz, 750Hz, 1000Hz, 1500Hz, 2000Hz and be each respectively provided with in the environment of voice and record 100 duration 30s respectively Effective sample speech data；2000 sample speech datas are divided into speech-sound data sequence the most respectively in units of 3s, Carry out 20000 elements of all speech-sound data sequences out of order being arranged to make up sample sequence, be successively read in sample sequence Element: for each element, remembers that the ensemble average decibel value of this element is x1, overall code check be x2, this element is carried out frequency domain Analyze, with 50Hz as starting point, calculate the rate of change often crossing 50Hz decibel value, record first rate of change frequency more than 0.1dB/Hz Rate value is x3, first rate of change frequency values less than-0.1dB/Hz be x4If not finding qualified x3, then by x3Set For 50Hz, if not finding qualified x4, then by x4It is set as 1200Hz, calculates x3To x4The average decibels of frequency range is x5, By x1、x2、x3、x4And x5As one group of training sample input data；20000 groups of training sample input data are combined each pantogen Corresponding invalidating expected outcome, to BP neural metwork training, keeps during training WithPerseverance is 0.
Further: described main control chip is also associated with FM module.
Further: described main control chip is also associated with display screen.
Further: described main control chip is also associated with memory module.
Relative to prior art, the invention have the advantages that (1) present invention has discrimination module based on FPGA, energy Whether enough utilize the neural network algorithm trained to determine according to the characteristic information of speech data fragment is that voice content is blank Invalid voice data slot, and invalid voice data slot is replaced with blank sound data, have compressed transmitting audio data Size, reduces the calling extension set demand to bandwidth；(2) this detection method utilizes neutral net to differentiate speech data, There is the advantage that None-linear approximation ability is strong, judging efficiency is high and accuracy rate is high；(3) neutral net introduces pretreatment layer, Owing to everyone voice frequency range is more concentrated, therefore part flexible strategy are carried out by pretreatment layer forcing to set, and will First rate of change is more than the frequency values x of 0.1dB/Hz3With the frequency values x that first rate of change is less than-0.1dB/Hz4Both Dependency is more apparent but that cannot be completely integrated characteristic information has carried out incomplete merging treatment, the most again by pretreatment The result of layer exports in intermediate layer, it is ensured that x during follow-up calculating3And x4All the time possess certain dependency, thus carry The high accuracy of judged result, also improves the efficiency of training simultaneously；(4) activation primitive of pretreatment layer sets and takes into full account X3And x4Two incomplete merging treatment of characteristic information are in terms of computational efficiency, differential solve difficulty and dependency reservation Requirement, have solve, training effectiveness is high and judgment accuracy is high advantage.
Accompanying drawing explanation
Fig. 1 is the structural representation of Multifunctional calling extension set proposed by the invention.
Fig. 2 is the structural representation of BP neutral net.
Detailed description of the invention
Describe technical scheme below in conjunction with the accompanying drawings in detail:
Such as Fig. 1, Multifunctional calling extension set, including main control chip, also include the wheat being each connected respectively with main control chip Gram wind, speaker and receiver module, described main control chip connects also by level cache discrimination module, described discrimination module Connected by L2 cache and have sending module；Described main control chip is also associated with FM module, LCDs and memory module, institute State memory module can be built-in storage chip can also be to be accessed the flash disk device of this calling extension set by USB interface.
Patient can pass through this calling extension set and host communication, it is also possible to listens to FM by speaker and broadcasts, or appreciates storage Music in module.LCDs may also display the relevant informations such as the name of patient, age.
Described discrimination module, based on fpga chip, has been built in described fpga chip for differentiating invalid voice data slot BP neutral net, described BP neutral net differentiate invalid voice data slot method step be:
(A) voice that mike is included by main control chip is converted into speech data, it is contemplated that voice frequency range be 65 to 1100Hz, all filters the frequency range of below 50Hz and more than 1200Hz in this speech data, then the speech data after filtering In units of 3s, it is divided into speech-sound data sequence and this speech-sound data sequence is left in level cache；
Main control chip is successively read the element in speech-sound data sequence from level cache, is handled as follows respectively:
(A-1) the ensemble average decibel value remembering this element is x1, overall code check be x2；
(A-2) this element is carried out frequency-domain analysis, with 50Hz as starting point, calculate the rate of change often crossing 50Hz decibel value, note Recording first rate of change frequency values more than 0.1dB/Hz is x3, first rate of change frequency values less than-0.1dB/Hz be x4； If not finding qualified x3, then by x3It is set as 50Hz, if not finding qualified x4, then by x4It is set as 1200Hz；
(A-3) x is calculated3To x4The average decibel value of frequency range is x5；
(A-4) by x1、x2、x3、x4And x5Store to level cache as one group of input data；
(B) it is sequentially delivered in the BP neutral net of discrimination module sentence by each group input data in level cache Not；
This BP neutral net is disposed with input layer, pretreatment layer, intermediate layer and output layer along input to outbound course；
Described input layer includes for inputting x1Input block one, be used for inputting x2Input block two, be used for inputting x3 Input block three, be used for inputting x4Input block four and be used for inputting x5Input block five；
Described pretreatment layer includes pretreatment unit one, pretreatment unit two, pretreatment unit three and pretreatment unit four；
Described intermediate layer includes temporary location one, temporary location two and temporary location three；
Described output layer includes output unit；
Described input layer, pretreatment layer, intermediate layer and output layer are respectively the 1st layer of BP neutral net, the 2nd layer, the 3rd layer With the 4th layer；
Described input block one, input block two, input block three, input block four and input block five are respectively the 1st Unit the 1st, Unit the 2nd, Unit the 3rd, Unit the 4th and Unit the 5th of layer；
Described pretreatment unit one, pretreatment unit two, pretreatment unit three and pretreatment unit four are respectively the 2nd layer Unit the 1st, Unit the 2nd, Unit the 3rd and Unit the 4th；
Described temporary location one, temporary location two and temporary location three are respectively Unit the 1st, Unit the 2nd and the of the 3rd layer Unit 3；
Described output unit is Unit the 1st of the 4th layer；
If the output valve of l layer i-th cell isBias term isActivation primitive is fi (l)(), the unit of l layer Sum is n(l), the output valve of l layer jth unitWeights when being transferred to l+1 layer i-th cell are
Then for the 1st layer:
For the 2nd to 4 layer:
IfWithPerseverance is 0, and this reason arranged is: each The voice frequency range of people is typically more concentrated, i.e. x3With x4Difference will not be excessive, therefore in pretreatment layer carries out part flexible strategy Force to set, and by x3And x4Both dependencys are more apparent but that cannot be completely integrated characteristic information has been carried out non-fully The merging treatment of property, exports in intermediate layer by the result of pretreatment layer, it is ensured that x during follow-up calculating the most again3And x4 All the time possess certain dependency, thus improve the accuracy of judged result, also improve the efficiency of training simultaneously；
According to the data of input, BP neutral net judges whether this element is invalid voice data slot, if invalid language This element is then replaced with blank sound data by sound data slot；
(C) speech-sound data sequence that replacement was processed by discrimination module is sent in L2 cache.
The activation primitive of described pretreatment layer each unit is:
f i ( l ) ( x ) = e 3 x + 3 x 3 - 1 e 2 x + 1 - x , x &GreaterEqual; 0 sin ( x 5 ) , x < 0 .
The setting of this activation primitive has taken into full account x3And x4Effect is being calculated after two incomplete merging treatment of characteristic information Rate, differential solve the requirement in terms of difficulty and dependency reservation, have solve, that training effectiveness is high and judgment accuracy is high is excellent Point；
The activation primitive of described intermediate layer and output layer each unit is:
fi (l)(x)=max (0, x-0.2).
The training method of this BP neutral net is: background noise frequency be 20Hz, 80Hz, 150Hz, 200Hz, 360Hz, 500Hz, 750Hz, 1000Hz, 1500Hz, 2000Hz and each all without recording 100 duration 30s' in the environment of voice respectively Invalid sample voice data, and background noise frequency be 20Hz, 80Hz, 150Hz, 200Hz, 360Hz, 500Hz, 750Hz, 1000Hz, 1500Hz, 2000Hz and be each respectively provided with the effective sample language recording 100 duration 30s in the environment of voice respectively Sound data；During sampling, different environment should be selected as far as possible, and the personnel arranging sound ray mutually different participate in the record of voice System；2000 sample speech datas are divided into speech-sound data sequence the most respectively in units of 3s, by all speech data sequences 20000 elements of row carry out out of order being arranged to make up sample sequence, the element being successively read in sample sequence: for every unitary Element, remembers that the ensemble average decibel value of this element is x1, overall code check be x2, this element is carried out frequency-domain analysis, with 50Hz for rising Point, calculates the rate of change often crossing 50Hz decibel value, and recording first rate of change frequency values more than 0.1dB/Hz is x3, first The rate of change frequency values less than-0.1dB/Hz is x4If not finding qualified x3, then by x3It is set as 50Hz, if not finding Qualified x4, then by x4It is set as 1200Hz, calculates x3To x4The average decibels of frequency range is x5, by x1、x2、x3、x4And x5 As one group of training sample input data；20000 groups of training sample input data are combined the effective/nothing corresponding to each pantogen Effect expected outcome, to BP neural metwork training, keeps during trainingWithPerseverance is 0.
Patient, due to reasons such as health status, occurs that in communication the sound of stage rests often, thus the voice of transmission Data volume is relatively big, occupies substantial portion of bandwidth.In order to the fragment that voice content in speech data is blank is removed, with Reducing the demand to bandwidth, this programme is by using discrimination module, and it is invalid to determine according to the characteristic information of speech data fragment Speech data fragment, is led to by L2 cache after then invalid voice data slot replacing with the blank sound data that volume is minimum again Cross sending module to transmit to main frame, have compressed the size of transmitting audio data, reduce the calling extension set demand to bandwidth.

Claims (7)

1. Multifunctional calling extension set, including main control chip, also includes the mike being each connected with main control chip respectively, raises one's voice Device and receiver module, it is characterised in that: described main control chip connects also by level cache discrimination module, described differentiation mould Block is connected by L2 cache sending module；
Described discrimination module, based on fpga chip, builds the BP for differentiating invalid voice data slot in described fpga chip Neutral net, described BP neutral net differentiates that the method step of invalid voice data slot is:
(A) voice that mike is included by main control chip is converted into speech data, by below 50Hz in this speech data and The frequency range of more than 1200Hz all filters, then the speech data after filtering is divided into speech-sound data sequence will in units of 3s This speech-sound data sequence leaves in level cache；
Main control chip is successively read the element in speech-sound data sequence from level cache, is handled as follows respectively:
(A-1) the ensemble average decibel value remembering this element is x1, overall code check be x2；
(A-2) this element is carried out frequency-domain analysis, with 50Hz as starting point, calculate the rate of change often crossing 50Hz decibel value, record the One rate of change frequency values more than 0.1dB/Hz is x3, first rate of change frequency values less than-0.1dB/Hz be x4；If not Find qualified x3, then by x3It is set as 50Hz, if not finding qualified x4, then by x4It is set as 1200Hz；
(A-3) x is calculated3To x4The average decibel value of frequency range is x5；
(A-4) by x1、x2、x3、x4And x5Store to level cache as one group of input data；
(B) it is sequentially delivered in the BP neutral net of discrimination module differentiate by each group input data in level cache；
This BP neutral net is disposed with input layer, pretreatment layer, intermediate layer and output layer along input to outbound course；
Described input layer includes for inputting x1Input block one, be used for inputting x2Input block two, be used for inputting x3Defeated Enter unit three, for inputting x4Input block four and be used for inputting x5Input block five；
Described pretreatment layer includes pretreatment unit one, pretreatment unit two, pretreatment unit three and pretreatment unit four；
Described intermediate layer includes temporary location one, temporary location two and temporary location three；
Described output layer includes output unit；
Described input layer, pretreatment layer, intermediate layer and output layer are respectively the 1st layer of BP neutral net, the 2nd layer, the 3rd layer and 4 layers；
Described input block one, input block two, input block three, input block four and input block five are respectively the 1st layer Unit the 1st, Unit the 2nd, Unit the 3rd, Unit the 4th and Unit the 5th；
Described pretreatment unit one, pretreatment unit two, pretreatment unit three and pretreatment unit four are respectively the 1st of the 2nd layer Unit, Unit the 2nd, Unit the 3rd and Unit the 4th；
Unit the 1st, Unit the 2nd and the 3rd that described temporary location one, temporary location two and temporary location three are respectively the 3rd layer are single Unit；
Described output unit is Unit the 1st of the 4th layer；
If the output valve of l layer i-th cell isBias term isActivation primitive isThe unit sum of l layer For n(l), the output valve of l layer jth unitWeights when being transferred to l+1 layer i-th cell are
Then for the 1st layer:
For the 2nd to 4 layer:
IfWithPerseverance is 0；
According to the data of input, BP neutral net judges whether this element is invalid voice data slot, if invalid voice number Then this element is replaced with blank sound data according to fragment；
(C) speech-sound data sequence that replacement was processed by discrimination module is sent in L2 cache.
2. Multifunctional calling extension set as claimed in claim 1, it is characterised in that: the activation primitive of described pretreatment layer each unit For:
f i ( l ) ( x ) = e 3 x + 3 x 3 - 1 e 2 x + 1 - x , x &GreaterEqual; 0 sin ( x 5 ) , x < 0 .
3. Multifunctional calling extension set as claimed in claim 1, it is characterised in that: swashing of described intermediate layer and output layer each unit Function of living is:
fi (l)(x)=max (0, x-0.2).
4. Multifunctional calling extension set as claimed in claim 1, it is characterised in that the training method of BP neutral net is: in background Noise frequency is 20Hz, 80Hz, 150Hz, 200Hz, 360Hz, 500Hz, 750Hz, 1000Hz, 1500Hz, 2000Hz and each All without the environment of voice is recorded the invalid sample voice data of 100 duration 30s respectively, and in background noise frequency it is 20Hz, 80Hz, 150Hz, 200Hz, 360Hz, 500Hz, 750Hz, 1000Hz, 1500Hz, 2000Hz and be each respectively provided with voice Environment in record the effective sample speech data of 100 duration 30s respectively；By 2000 sample speech datas the most respectively with 3s is that unit is divided into speech-sound data sequence, carries out 20000 elements of all speech-sound data sequences out of order being arranged to make up sample This sequence, the element being successively read in sample sequence: for each element, remember that the ensemble average decibel value of this element is x1, whole Body code check is x2, this element is carried out frequency-domain analysis, with 50Hz as starting point, calculates the rate of change often crossing 50Hz decibel value, record First rate of change frequency values more than 0.1dB/Hz is x3, first rate of change frequency values less than-0.1dB/Hz be x4If, Do not find qualified x3, then by x3It is set as 50Hz, if not finding qualified x4, then by x4It is set as 1200Hz, meter Calculate x3To x4The average decibels of frequency range is x5, by x1、x2、x3、x4And x5As one group of training sample input data；By 20000 groups Training sample input data combine the invalidating expected outcome corresponding to each pantogen to BP neural metwork training, during training KeepWithPerseverance is 0.
5. the Multifunctional calling extension set as described in Claims 1-4 is arbitrary, it is characterised in that: described main control chip is also associated with FM module.
6. the Multifunctional calling extension set as described in Claims 1-4 is arbitrary, it is characterised in that: described main control chip is also associated with Display screen.
7. the Multifunctional calling extension set as described in Claims 1-4 is arbitrary, it is characterised in that: described main control chip is also associated with Memory module.
CN201610670271.7A 2016-08-15 2016-08-15 Multifunctional calling extension set Active CN106297771B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610670271.7A CN106297771B (en) 2016-08-15 2016-08-15 Multifunctional calling extension set

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610670271.7A CN106297771B (en) 2016-08-15 2016-08-15 Multifunctional calling extension set

Publications (2)

Publication Number Publication Date
CN106297771A true CN106297771A (en) 2017-01-04
CN106297771B CN106297771B (en) 2019-07-05

Family

ID=57671418

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610670271.7A Active CN106297771B (en) 2016-08-15 2016-08-15 Multifunctional calling extension set

Country Status (1)

Country Link
CN (1) CN106297771B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110418245A (en) * 2018-04-28 2019-11-05 深圳市冠旭电子股份有限公司 A kind of method, apparatus and terminal device reducing Baffle Box of Bluetooth response delay

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103458056A (en) * 2013-09-24 2013-12-18 贵阳世纪恒通科技有限公司 Speech intention judging method based on automatic classification technology for automatic outbound system
CN105393302A (en) * 2013-07-17 2016-03-09 三星电子株式会社 Multi-level speech recognition
CN105489217A (en) * 2015-12-30 2016-04-13 武传胜 Method and system of monitoring the aged through sound by means of internet of things

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105393302A (en) * 2013-07-17 2016-03-09 三星电子株式会社 Multi-level speech recognition
CN103458056A (en) * 2013-09-24 2013-12-18 贵阳世纪恒通科技有限公司 Speech intention judging method based on automatic classification technology for automatic outbound system
CN105489217A (en) * 2015-12-30 2016-04-13 武传胜 Method and system of monitoring the aged through sound by means of internet of things

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
张晨昊: "基于CTI技术的中小企业呼叫中心的设计与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110418245A (en) * 2018-04-28 2019-11-05 深圳市冠旭电子股份有限公司 A kind of method, apparatus and terminal device reducing Baffle Box of Bluetooth response delay
CN110418245B (en) * 2018-04-28 2021-03-19 深圳市冠旭电子股份有限公司 Method and device for reducing reaction delay of Bluetooth sound box and terminal equipment

Also Published As

Publication number Publication date
CN106297771B (en) 2019-07-05

Similar Documents

Publication Publication Date Title
CN103026738B (en) The method of signal transacting and hearing aid device system in hearing aid device system
CN102652337B (en) Device and method for acoustic communication
CN107569236B (en) Multifunctional hearing test and hearing aid system and hearing test method thereof
Chen et al. Predicting the intelligibility of vocoded and wideband Mandarin Chinese
CN101716108B (en) A cochlear electronic signal processing and electrical stimulation coding system
CN106340309A (en) Dog bark emotion recognition method and device based on deep learning
CN110544532A (en) An APP-based sound source spatial localization ability detection system
CN106303874B (en) A kind of adaptive confirmed method of completing the square of digital deaf-aid
CN108357445A (en) Car masking sound quality self-adapting control system and method
US20120008800A1 (en) Telephone enhancements
CN102467906A (en) Hearing detection method and system
Zhu et al. Contribution of modulation spectral features on the perception of vocal-emotion using noise-vocoded speech
Unoki et al. Relationship between contributions of temporal amplitude envelope of speech and modulation transfer function in room acoustics to perception of noise-vocoded speech
CN110825227A (en) Ward intelligent monitoring system and method based on VR technology
Grant et al. Integration efficiency for speech perception within and across sensory modalities by normal-hearing and hearing-impaired individuals
CN106297771A (en) Multifunctional calling extension set
Luo et al. Design of embedded real-time system for snoring and OSA detection based on machine learning
Patil et al. Marathi speech intelligibility enhancement using I-AMS based neuro-fuzzy classifier approach for hearing aid users
CN103099625B (en) Test method and system for auditory time modulation transfer function
CN111091884A (en) Patient main index matching system and method of hospital information integration platform
CN1668058A (en) Recursive least square difference based subband echo canceller
KR100669532B1 (en) USB Digital Stethoscope
Sanchez-Lopez et al. Auditory tests for characterizing hearing deficits: The BEAR test battery
CN105411535B (en) A kind of device and method that tinnitus treatment effect is assessed using position location technology
CN115553760A (en) Music synthesis method for tinnitus rehabilitation and online tinnitus diagnosis and rehabilitation system

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant
CP01 Change in the name or title of a patent holder
CP01 Change in the name or title of a patent holder

Address after: No.8 cheshan Road, Zhifu District, Yantai City, Shandong Province

Patentee after: Huasheng Medical Technology Co.,Ltd.

Address before: No.8 cheshan Road, Zhifu District, Yantai City, Shandong Province

Patentee before: YANTAI BEACON MEDICAL TECHNOLOGY Co.,Ltd.

PE01 Entry into force of the registration of the contract for pledge of patent right
PE01 Entry into force of the registration of the contract for pledge of patent right

Denomination of invention: Multi function call extension

Effective date of registration: 20210708

Granted publication date: 20190705

Pledgee: Rizhao Bank Co.,Ltd. Yantai Zhifu sub branch

Pledgor: Huasheng Medical Technology Co.,Ltd.

Registration number: Y2021980005938