CN109597305A - Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis - Google Patents
Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis Download PDFInfo
- Publication number
- CN109597305A CN109597305A CN201811463077.7A CN201811463077A CN109597305A CN 109597305 A CN109597305 A CN 109597305A CN 201811463077 A CN201811463077 A CN 201811463077A CN 109597305 A CN109597305 A CN 109597305A
- Authority
- CN
- China
- Prior art keywords
- music signal
- analysis
- clothes
- real
- music
- 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
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Electrically Operated Instructional Devices (AREA)
Abstract
Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis the present invention relates to a kind of characterized by comprising data pre-acquired equipment；Establishing in a data storage device has disaggregated model；Music signal processing equipment；Control system controls shock module according to the real-time vibration information received, so that the input module for being worn on user's difference physical feeling is issued in the corresponding time according to the beat of real-time music signal and shaken, the corresponding physical feeling of user is reminded to make movement.The present invention is combined using data analysis equipment with shake control system, intelligently solves the prompting function of music and dance, and the human cost of associated mechanisms can be made to substantially reduce.The learning cost of the equipment is low, and target user is extensive.The equipment uses cloud data storage device, and Information Security is high, widely used, and ordinary user crowd can also be facilitated to learn dancing technical ability.
Description
Technical field
The intelligent clothing vibration reminding system based on music signal processing and big data analysis that the present invention relates to a kind of, belongs to
Data analysis, internet and field of signal processing.
Background technique
As the improvement of people's living standards, the rhythm of people's lives is getting faster, many people wish learning some industry
The time can be saved when remaining hobby.Some dancing fans wish to accelerate the progress of dancing study simultaneously.However it lacks at present
Lack such equipment to meet these demands.
Summary of the invention
The purpose of the present invention is: the movement prompting of each beat is provided to be unfamiliar with the dancer of song, dancer is facilitated to jump
It accurately acts accordingly when dance, can be also used for the dancing study of the masses.
In order to achieve the above object, Speech Signal Analysis and big data are based on the technical solution of the present invention is to provide a kind of
The clothes of analysis shake intelligent reminding system characterized by comprising
Data pre-acquired equipment extracts the music data in audio database and inputs vibration corresponding with music data
Training dataset is formed after information inputs to data storage device；
Establishing in a data storage device has disaggregated model, and disaggregated model is trained using training data, is trained
Disaggregated model afterwards；
Music signal processing equipment, for extracting the music signal characteristic of externally input real-time music signal, and
Music signal characteristic is uploaded to data storage device, by the disaggregated model after the training in data storage device according to connecing
The data received are predicted to obtain corresponding real-time vibration information, and real-time vibration information is fed back to music signal processing equipment,
Music signal processing equipment exports the real-time vibration information received to control system；
Control system controls shock module according to the real-time vibration information received, so that being worn on user's difference body
The input module of position, which is issued according to the beat of real-time music signal in the corresponding time, to be shaken, and the corresponding physical feeling of user is reminded
Make movement.
Preferably, externally input real-time music signal inputs to institute after being converted to digital signal by A/D conversion equipment
State music signal processing equipment.
Preferably, the music signal processing equipment is uploaded to after carrying out dimension-reduction treatment to the music signal characteristic
The data storage device.
Preferably, the disaggregated model selects random forest sorting algorithm, and the random forest for carrying out model training selection is calculated
The parameter of method includes that the number of tree is 100, and the depth of tree is 6, and constituting the algorithm that the decision tree of random forest uses is C4.5, with
Machine forest model not beta pruning, while improving using cross validation method the generalization ability of algorithm.
Preferably, user's difference physical feeling includes arm input module, neck input module, leg input module and chest
Portion's input module.
It preferably, further include feedback self learning system, feedback self learning system has user's input port, for recording user
Think to shake the time and be spaced inappropriate point, and is modified accordingly, then be passed to the data storage device.
Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis the present invention provides a kind of,
It mainly has the beneficial effect that the present invention is combined using data analysis equipment with shake control system, intelligently solves
The prompting function of music and dance can be such that the human cost of associated mechanisms substantially reduces.The learning cost of the equipment is low, is applicable in people
Group is extensively.The equipment uses cloud data storage device, and Information Security is high, widely used, can also facilitate ordinary user people
Group's study dancing technical ability.There are also self feed back learning systems for the equipment, can constantly enhance the performance of model, improve classification results
Accuracy.
Detailed description of the invention
Fig. 1 is provided by the invention a kind of based on Speech Signal Analysis and the clothes of big data analysis vibration intelligent reminding system
The system block diagram of system.
Specific embodiment
Present invention will be further explained below with reference to specific examples.It should be understood that these embodiments are merely to illustrate the present invention
Rather than it limits the scope of the invention.In addition, it should also be understood that, after reading the content taught by the present invention, those skilled in the art
Member can make various changes or modifications the present invention, and such equivalent forms equally fall within the application the appended claims and limited
Range.
As shown in Figure 1, the present invention provides a kind of based on music signal processing and the clothes of big data analysis vibration intelligence
System for prompting, including A/D conversion equipment 1 and music signal processing equipment 2.A/D conversion equipment is furnished with music signal input port,
Music signal processing equipment 2 is connected with data storage device 3 and control system 4.The input terminal of data storage device 3 also with data
Pre-acquired system 5 is connected.Data pre-acquired system 5 is connected with audio database 6 and vibration information input system 7.Control system 4
Output end be connected with shock module 8, and connect with the shock module 8 there are also arm input module 9, neck input module
10, leg input module 11 and foot's input module 12.Feedback self learning system 13 has user's input port, for recording user
Think to shake the time and be spaced inappropriate point, and modified accordingly, then incoming storage equipment.
It is provided by the invention a kind of based on music signal processing and the clothes of big data analysis vibration intelligent reminding system, lead to
The analysis for crossing the characteristic using machine learning model and to music signal analyzes the music signal of input in some time
The possible vibration information of point, to remind wearer to wave some position of body at the time point, so that it is ripe to improve wearer
Know the speed of music and dance.Its key step is as follows:
Step 1: the processing of music signal
The music signal can be analog signal or digital signal, be number if the music signal downloaded from network
Signal needs not move through A/D conversion equipment 1 at this time；If music signal need to be converted from the real-time input of sound into A/D is crossed
Equipment 1 is processed into digital signal.Simultaneously to the music signal carry out Signal Pretreatment, including preemphasis, end-point detection, adding window and
Sub-frame processing.The music signal processing equipment 2 receives the digital signal that A/D conversion equipment 1 exports, and is believed using built-in music
The features such as number feature extracting method such as amplitude energy, fundamental frequency, formant and mel-frequency cepstrum coefficient MFCC obtain special
It needs to carry out dimension-reduction treatment, including linear dimension reduction method and Method of Nonlinear Dimensionality Reduction to characteristic after sign.Linear dimension reduction method has
Principal component analysis (PCA), linear discriminant analysis (LDA), Method of Nonlinear Dimensionality Reduction is mainly manifold learning arithmetic, for reducing unrelated
The quantity of feature, to promote the operational efficiency and recognition accuracy of sorting algorithm.
Step 2: data pre-acquired and model training
The step by Berlin musical database extract data, be input to the data pre-acquired system 5, selection
The data characteristics of Berlin musical database includes amplitude energy, fundamental frequency, formant and mel-frequency cepstrum coefficient
The features such as MFCC, while corresponding vibration information is inputted, random forest sorting algorithm is selected inside data storage device 3, is carried out
The parameter for the random forests algorithm that model training is chosen includes that the number of tree is 100, and the depth of tree is 6, constitutes random forest
The algorithm that decision tree uses is C4.5, Random Forest model not beta pruning, while the general of algorithm is improved using cross validation method
Change ability.
Step 3: output prediction of result
The output result of the music signal processing equipment 2 uploads to data storage device 3, carries out model measurement, output
Classification results, output result are the vibration information of the music signal of input, and the result is downloaded to music signal processing equipment
2, the vibration information of output is burnt to control system 4 again by music signal processing equipment 2, and the control system 4 is generally monolithic
General 51 single-chip microcontroller can be used in machine system, and the single-chip computer control system of ARM framework, programming language can be used to improve performance
C language is generally used, operation result is transferred to the shock module 8 being connected with control system 4 after the completion of burning.
Step 4: vibration output
The shock module 8 is connected with arm input module 9, neck input module 10, leg input module 11 and foot
Input module 12, the module are generally made of micro-vibration-motor TELESKY 1027, and the micro motor voltage rating is
3V, starting current 90mA, there are two pins, are connected respectively with the low level end of control system 4 and high level end, the vibration
The vibration information that module 8 receives control system 4 starts to shake after vibration information is transmitted to module 9,10,11,12.The vibration
Information be by model prediction as a result, vibration reminding can be given when user forget to touch movement, so that improving makes
The learning efficiency of user.
Claims (6)
1. a kind of shake intelligent reminding system based on Speech Signal Analysis and the clothes of big data analysis characterized by comprising
Data pre-acquired equipment (5) extracts the music data in audio database (6) and inputs shake corresponding with music data
Training dataset, which is formed, after dynamic information inputs to data storage device (3)；
Establishing in data storage device (3) has disaggregated model, and disaggregated model is trained using training data, after being trained
Disaggregated model；
Music signal processing equipment (2), for extracting the music signal characteristic of externally input real-time music signal, and will
Music signal characteristic is uploaded to data storage device (3), by the disaggregated model root after the training in data storage device (3)
It predicts to obtain corresponding real-time vibration information according to the data received, and real-time vibration information is fed back into music signal processing and is set
Standby (2), music signal processing equipment (2) export the real-time vibration information received to control system (4)；
Control system (4) is according to real-time vibration information control shock module (8) received, so that being worn on user's difference body
The input module at position, which is issued according to the beat of real-time music signal in the corresponding time, to be shaken, and the corresponding body of user is reminded
Make movement in position.
2. as described in claim 1 a kind of based on Speech Signal Analysis and the clothes of big data analysis vibration intelligent reminding system
System, which is characterized in that externally input real-time music signal inputs to after being converted to digital signal by A/D conversion equipment (1)
The music signal processing equipment (2).
3. as described in claim 1 a kind of based on Speech Signal Analysis and the clothes of big data analysis vibration intelligent reminding system
System, which is characterized in that the music signal processing equipment (2) uploads after carrying out dimension-reduction treatment to the music signal characteristic
To the data storage device (3).
4. as described in claim 1 a kind of based on Speech Signal Analysis and the clothes of big data analysis vibration intelligent reminding system
System, which is characterized in that the disaggregated model selects random forest sorting algorithm, carries out the random forests algorithm of model training selection
Parameter include the number of tree be 100, the depth of tree is 6, and constituting the algorithm that the decision tree of random forest uses is C4.5, at random
Forest model not beta pruning, while improving using cross validation method the generalization ability of algorithm.
5. as described in claim 1 a kind of based on Speech Signal Analysis and the clothes of big data analysis vibration intelligent reminding system
System, which is characterized in that user's difference physical feeling includes arm input module (9), neck input module (10), leg input mould
Block (11) and chest input module (12).
6. as described in claim 1 a kind of based on Speech Signal Analysis and the clothes of big data analysis vibration intelligent reminding system
System, which is characterized in that further include feedback self learning system (13), feedback self learning system (13) has user's input port, is used for
Record user thinks to shake the time and is spaced inappropriate point, and is modified accordingly, then be passed to the data storage device
(3)。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811463077.7A CN109597305A (en) | 2018-12-03 | 2018-12-03 | Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811463077.7A CN109597305A (en) | 2018-12-03 | 2018-12-03 | Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109597305A true CN109597305A (en) | 2019-04-09 |
Family
ID=65959452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811463077.7A Pending CN109597305A (en) | 2018-12-03 | 2018-12-03 | Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109597305A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008151642A1 (en) * | 2007-06-12 | 2008-12-18 | Nokia Corporation | Directing shoe insole |
| CN202601030U (en) * | 2012-05-11 | 2012-12-12 | 长春大学 | Deaf-mute dance training and stage prompting system based on zigbee technology |
| CN104882144A (en) * | 2015-05-06 | 2015-09-02 | 福州大学 | Animal voice identification method based on double sound spectrogram characteristics |
| CN204670479U (en) * | 2015-04-15 | 2015-09-30 | 温州芳植生物科技有限公司 | A kind of dancing shoe of improvement |
| US20150332659A1 (en) * | 2014-05-16 | 2015-11-19 | Not Impossible LLC | Sound vest |
| CN105405337A (en) * | 2015-10-22 | 2016-03-16 | 小天才科技有限公司 | Auxiliary music playing method and system |
| CN205385902U (en) * | 2016-03-14 | 2016-07-20 | 温州职业技术学院 | Multi -functional high -heeled shoes for dance |
| CN106707823A (en) * | 2015-07-24 | 2017-05-24 | 时晓欣 | Wireless-communication-based square dancing system formed by console and intelligent upper outer garment |
-
2018
- 2018-12-03 CN CN201811463077.7A patent/CN109597305A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2008151642A1 (en) * | 2007-06-12 | 2008-12-18 | Nokia Corporation | Directing shoe insole |
| CN202601030U (en) * | 2012-05-11 | 2012-12-12 | 长春大学 | Deaf-mute dance training and stage prompting system based on zigbee technology |
| US20150332659A1 (en) * | 2014-05-16 | 2015-11-19 | Not Impossible LLC | Sound vest |
| CN204670479U (en) * | 2015-04-15 | 2015-09-30 | 温州芳植生物科技有限公司 | A kind of dancing shoe of improvement |
| CN104882144A (en) * | 2015-05-06 | 2015-09-02 | 福州大学 | Animal voice identification method based on double sound spectrogram characteristics |
| CN106707823A (en) * | 2015-07-24 | 2017-05-24 | 时晓欣 | Wireless-communication-based square dancing system formed by console and intelligent upper outer garment |
| CN105405337A (en) * | 2015-10-22 | 2016-03-16 | 小天才科技有限公司 | Auxiliary music playing method and system |
| CN205385902U (en) * | 2016-03-14 | 2016-07-20 | 温州职业技术学院 | Multi -functional high -heeled shoes for dance |
Non-Patent Citations (1)
| Title |
|---|
| 黄鸿铿 等: "用Bark频谱投影识别低信噪比动物声音", 《智能系统学报》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Nam et al. | Deep learning for audio-based music classification and tagging: Teaching computers to distinguish rock from bach | |
| Kim et al. | Towards speech emotion recognition" in the wild" using aggregated corpora and deep multi-task learning | |
| EP3803846B1 (en) | Autonomous generation of melody | |
| US11511436B2 (en) | Robot control method and companion robot | |
| Tahon et al. | Towards a small set of robust acoustic features for emotion recognition: challenges | |
| Zhang et al. | Cooperative learning and its application to emotion recognition from speech | |
| Ntalampiras et al. | Modeling the temporal evolution of acoustic parameters for speech emotion recognition | |
| CN109523994A (en) | A kind of multitask method of speech classification based on capsule neural network | |
| CN105761720A (en) | Interaction system based on voice attribute classification, and method thereof | |
| Li et al. | Speech emotion recognition using 1d cnn with no attention | |
| Song et al. | Music auto-tagging using deep recurrent neural networks | |
| Nathan et al. | EMOSIC—An emotion based music player for Android | |
| Garg et al. | Speech based Emotion Recognition based on hierarchical decision tree with SVM, BLG and SVR classifiers | |
| US10593319B1 (en) | Parallelization of instruction steps | |
| CN109741724A (en) | Make the method, apparatus and intelligent sound of song | |
| CN113813609A (en) | Game music style classification method and device, readable medium and electronic equipment | |
| Cao et al. | Speaker-independent speech emotion recognition based on random forest feature selection algorithm | |
| Mande et al. | EMOTION DETECTION USING AUDIO DATA SAMPLES. | |
| CN111754981A (en) | Command word recognition method and system using mutual prior constraint model | |
| Gao et al. | A novel music emotion recognition model for scratch-generated music | |
| CN109597305A (en) | Intelligent reminding system is shaken based on Speech Signal Analysis and the clothes of big data analysis | |
| Zhang et al. | Emotion recognition in speech using multi-classification SVM | |
| Cummins et al. | Latest advances in computational speech analysis for mobile sensing | |
| Mangalam et al. | Emotion Recognition from Mizo Speech: A Signal Processing Approach | |
| He et al. | Automatic generation algorithm analysis of dance movements based on music–action association |
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 |
Application publication date: 20190409 |
|
| RJ01 | Rejection of invention patent application after publication |