CN101382590A - Sound source direction recognizing method, recognizing device and robot with the device - Google Patents
Sound source direction recognizing method, recognizing device and robot with the device Download PDFInfo
- Publication number
- CN101382590A CN101382590A CNA2007101212731A CN200710121273A CN101382590A CN 101382590 A CN101382590 A CN 101382590A CN A2007101212731 A CNA2007101212731 A CN A2007101212731A CN 200710121273 A CN200710121273 A CN 200710121273A CN 101382590 A CN101382590 A CN 101382590A
- Authority
- CN
- China
- Prior art keywords
- source direction
- sound source
- sound
- processor
- incident angle
- 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
Images
Landscapes
- Manipulator (AREA)
Abstract
The invention relates to a recognition method of the sound source direction, which comprises the steps of: collecting an aural signal by three sound collecting devices which are arranged at the three peaks of an equilateral triangle laid on one horizontal plane; transforming the aural signal to a digital signal by a voice signal processing device and conveying the digital signal to a processor; according to the time difference of the received digital signal, the processor obtains the incident angle of the sound source direction by inquiring a preset datasheet of the matchup of the time difference and the incident angle in an inquiring processor. The recognition device of sound source direction comprises: three sound collecting devices, a signal processing device and the processor. The robot with the voice recognition device of the invention comprises: a robot control system and a robot executing mechanism, and also a recognition device of the sound source direction; the recognition device is connected with the robot control system by an interface adapter. In the application area with lower accuracy demand, the invention achieves the goal of fast recognition on the sound source direction; furthermore, the invention has the advantages of lower facility cost, small volume, fast arithmetic speed and high recognition accuracy.
Description
Affiliated technical field
The present invention relates to a kind of artificial intelligence technology, relate in particular to a kind of sound source direction recognizing method, recognition device and have the robot of this device.
Background technology
At present, there are some ripe sound source direction discrimination technologies to be applied to the very high fields such as hydrospace detection of accuracy requirement usually, but the complexity of its algorithm, equipment cost is higher, it is transplanted in fields such as the toy manufacturing that same demand is arranged, robots, exist the difficult and high problem of manufacturing cost of coordinative composition of equipments, be difficult to realize good economic benefit.Simultaneously, fields such as toy manufacturing, robot do not have too high accuracy of identification requirement, so, need a kind of simple and practical sound source direction recognizing method and recognition device.
Robot turns to head Sounnd source direction to carry out the sound picked-up if can learn Sounnd source direction, and the people that can make experience robot and have life and friendliness like this, because people are aspectant when being accustomed to talking.For robot, when she forwards head to Sounnd source direction, can use the CCD pickup image, carry out status differentiation then, after status differentiation is intact, just carry out the specified work of people, it is an advanced person's intelligent robot that the robot that makes becomes.Equally, sound can interrupt the task that robot is carrying out, and loses no time to put down work on hand after hearing call, goes to people next door and goes to inquire that what can serve for people.Make people feel " come at someone's call " and sensation, therefore, sound source direction recognizing method, recognition device and have the robot of this device in field of artificial intelligence, have important use.
Summary of the invention
In order to overcome the deficiencies in the prior art, first purpose of the present invention is to provide a kind of simple and practical, cost is low, computing is fast sound source direction recognizing method, to reach under the not high situation of accuracy requirement, realizes the quick identification of sound source direction.
Second purpose of the present invention is to provide a kind of simple and practical, cost is low, computing is fast sound source direction recognition device, to reach under the not high situation of accuracy requirement, realizes the quick identification of sound source direction.
Second purpose of the present invention is to provide a kind of robot with sound source direction recognition device of the present invention, can be by of the fast effectively identification of sound source direction recognition device to sound source direction, and the subsequent action that is reached for robot provides the purpose of parameter.
The technical solution adopted for the present invention to solve the technical problems is:
A kind of sound source direction recognizing method comprises successively and carrying out:
Step S1: by three voice collection device collected sound signals of place, three summits of the equilateral triangle on surface level layout;
Step S2: described voice signal respectively by speech signal processing device, is treated to digital signal with it;
Step S3: described digital signal is transmitted in processor, and processor is according to the mistiming of receiving digital signals, by the mistiming of presetting in the query processor and the tables of data of incident angle corresponding relation, obtains the incident angle of sound source direction.
A kind of sound source direction recognition device comprises:
Three voice collection device are laid in three summits of the equilateral triangle of surface level, are used for collected sound signal;
Signal processing apparatus is used for voice signal is converted into digital signal;
Processor connects described signal processing apparatus by interface adapter, is used for the mistiming according to institute's receiving digital signals, by the mistiming of presetting in the query processor and the tables of data of incident angle corresponding relation, obtains the angle of sound source direction.
A kind of robot with speech recognition equipment, comprise robot control system and robot actuating mechanism, it is characterized in that: also comprise sound source direction recognition device as claimed in claim 5, described recognition device is connected in described robot control system by interface adapter.
The invention has the beneficial effects as follows that sound source direction recognizing method of the present invention and recognition device in the not high application of accuracy requirement, have reached the quick identification to sound source direction, its equipment cost is low, and volume is little, fast operation, recognition accuracy height.Robot of the present invention, because of it can carry out sound source direction identification, can be further and better finish the specified work of people, have higher performance, becoming is an advanced person's intelligent robot.
Description of drawings
Fig. 1 is the modular structure figure of sound source direction recognition device of the present invention.
Fig. 2 is a minimum identification angle key diagram in the embodiment of the invention.
Fig. 3 is the modular structure figure that the present invention has the robot of sound source direction recognition device.
Fig. 4 is the special sound discriminating direction modular structure figure in the embodiment of the invention.
Fig. 5 is the band pass filter circuit figure in the embodiment of the invention.
Fig. 6 is the peak-holding circuit circuit diagram in the embodiment of the invention.
Fig. 7 is the single channel sound source direction judging circuit figure in the embodiment of the invention.
Fig. 8 is the process flow diagram of sound source direction recognizing method embodiment of the present invention.
Embodiment
The present invention is further described below in conjunction with drawings and Examples.
The robot of embodiments of the invention is provided with source of sound recognition device of the present invention, and uses source of sound recognition methods of the present invention to carry out source of sound identification.
As shown in Figure 1, the source of sound recognition device of the embodiment of the invention comprises: three Mikes with pickup circuit are used for collected sound signal; Signal processing apparatus is used for voice signal is converted into digital signal; Single-chip microcomputer connects signal processing apparatus by interface adapter.Also comprise a serial line interface that is connected with robot control system.
As shown in Figure 2, three Mikes arranging of place, three summits of the equilateral triangle on a surface level of the robot of present embodiment.Described voice signal respectively by speech signal processing device, is treated to digital signal with it; Digital signal is transmitted in single-chip microcomputer, and single-chip microcomputer is according to the mistiming of receiving digital signals, by the mistiming of presetting in the query processor and the tables of data of incident angle corresponding relation, obtains the incident angle θ of sound source direction.
As shown in Figure 3, the robot with sound source direction recognition device of the embodiment of the invention, comprise robot control system and robot actuating mechanism, also comprise the sound source direction identification module of sound source direction recognition device embodiment of the present invention, the sound source direction identification module is connected in robot control system by interface adapter.
Reaching after generally, speech signal processing device utilizes monostable circuit to amplify set the threshold values voice signal and is shaped as digital signal.For special sound, can at first carry out spectrum analysis to specific sound by bandpass filter, determine the feature of sound, as shown in Figure 5, obtain being with logical frequency; By peak-holding circuit, as shown in Figure 6, filtered alternating signal is shaped as direct current signal then; By comparer, peak-holding circuit output conditioning is the digital signal of processor requirement at last.As shown in Figure 7, it is the hardware circuit that nonspecific sound source direction is differentiated, and in order to improve the robustness of system, this circuit need cooperate other software functions to use.
As shown in Figure 8, the sound source direction recognizing method of the embodiment of the invention comprises behind the single-chip microcomputer receiving digital signals and carries out flow process successively:
Receive first interrupt identification, then record interrupt the source voice collection device sequence number and start timer in the processor;
Receive second interrupt identification, it is poor then to write down the sequence number of the voice collection device of interrupting the source and obtain the timer very first time;
Receive interrupt identification for the third time, then the recording timing device is with respect to second mistiming of second interrupt identification;
By the mistiming of presetting in the query processor and the tables of data of incident angle corresponding relation, obtain the incident angle and the output of sound source direction.
In the present embodiment, as shown in Figure 2, mistiming and incident angle corresponding relation be, the cosine value of incident angle θ complementary angle multiply by the length of side X of equilateral triangle, and its result is that the minimum identification time and the velocity of sound are long-pending.In the present embodiment, the length of side of equilateral triangle is 15cm.
Above-described only is preferable possible embodiments of the present invention; described embodiment is not in order to limit scope of patent protection of the present invention; therefore the equivalent structure done of every utilization instructions of the present invention and accompanying drawing content changes, and in like manner all should be included in protection scope of the present invention.
Claims (10)
1, a kind of sound source direction recognizing method comprises successively and carrying out:
Step S1: by three voice collection device collected sound signals of place, three summits of the equilateral triangle on surface level layout;
Step S2: described voice signal respectively by speech signal processing device, is treated to digital signal with it;
Step S3: described digital signal is transmitted in processor, and processor is according to the mistiming of receiving digital signals, by the mistiming of presetting in the query processor and the tables of data of incident angle corresponding relation, obtains the incident angle of sound source direction.
2, sound source direction recognizing method as claimed in claim 1 is characterized in that: reaching after speech signal processing device described in the step S2 utilizes monostable circuit to amplify set the threshold values voice signal and is shaped as digital signal.
3, sound source direction recognizing method as claimed in claim 1 is characterized in that: for special sound, step S2 can be refined as,
Step S21: by bandpass filter specific sound is carried out spectrum analysis, determine the feature of sound, obtain being with logical frequency;
Step S22:, filtered alternating signal is shaped as direct current signal by peak-holding circuit;
Step S23:, peak-holding circuit output conditioning is the digital signal of processor requirement by comparer.
4, as claim 1 or 3 described sound source direction recognizing methods, it is characterized in that: described S3 comprises successively and carrying out:
Step S31: receive first interrupt identification, then record interrupt the source voice collection device sequence number and start timer in the processor;
Step S32: receive second interrupt identification, it is poor then to write down the sequence number of the voice collection device of interrupting the source and obtain the timer very first time;
Step S33: receive interrupt identification for the third time, then the recording timing device is with respect to second mistiming of second interrupt identification;
Step S34:, obtain the incident angle and the output of sound source direction by the mistiming of presetting in the query processor and the tables of data of incident angle corresponding relation.
5, sound source direction recognizing method as claimed in claim 1 is characterized in that, described mistiming and incident angle corresponding relation be, the cosine value of incident angle complementary angle multiply by the length of side of equilateral triangle, and its result is that the minimum identification time and the velocity of sound are long-pending.
6, a kind of sound source direction recognition device comprises:
Three voice collection device are laid in three summits of the equilateral triangle of surface level, are used for collected sound signal;
Signal processing apparatus is used for voice signal is converted into digital signal;
Processor connects described signal processing apparatus by interface adapter, is used for the mistiming according to institute's receiving digital signals, by the mistiming of presetting in the query processor and the tables of data of incident angle corresponding relation, obtains the angle of sound source direction.
7, sound source direction recognition device as claimed in claim 6 is characterized in that, described voice collection device is that to have be the Mike of pickup circuit, and the length of side of described equilateral triangle is 15cm.
8, sound source direction recognition device as claimed in claim 6 is characterized in that, also comprises a serial line interface that is connected with described processor.
9, sound source direction recognition device as claimed in claim 6 is characterized in that, described mistiming and incident angle corresponding relation be, the cosine value of incident angle complementary angle multiply by the length of side of equilateral triangle, and its result is that the minimum identification time and the velocity of sound are long-pending.
10, a kind of robot with speech recognition equipment, comprise robot control system and robot actuating mechanism, it is characterized in that: also comprise sound source direction recognition device as claimed in claim 6, described recognition device is connected in described robot control system by interface adapter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2007101212731A CN101382590A (en) | 2007-09-03 | 2007-09-03 | Sound source direction recognizing method, recognizing device and robot with the device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CNA2007101212731A CN101382590A (en) | 2007-09-03 | 2007-09-03 | Sound source direction recognizing method, recognizing device and robot with the device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101382590A true CN101382590A (en) | 2009-03-11 |
Family
ID=40462534
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CNA2007101212731A Pending CN101382590A (en) | 2007-09-03 | 2007-09-03 | Sound source direction recognizing method, recognizing device and robot with the device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101382590A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105068782A (en) * | 2015-08-11 | 2015-11-18 | 胡达广 | Recognition system for voice instruction of robot |
CN105807261A (en) * | 2016-04-26 | 2016-07-27 | 深思考人工智能机器人科技(北京)有限公司 | Device and method for distinguishing position by listening to sound |
CN106378786A (en) * | 2016-11-30 | 2017-02-08 | 北京百度网讯科技有限公司 | Robot based on artificial intelligence |
CN107004413A (en) * | 2014-11-28 | 2017-08-01 | 微软技术许可有限责任公司 | Expanding digital personal assistant acts supplier |
CN109831709A (en) * | 2019-02-15 | 2019-05-31 | 杭州嘉楠耘智信息科技有限公司 | Sound source orientation method and device and computer readable storage medium |
CN110193834A (en) * | 2019-05-16 | 2019-09-03 | 深圳市百思智能科技有限公司 | A kind of intelligent tracking system based on speech recognition |
-
2007
- 2007-09-03 CN CNA2007101212731A patent/CN101382590A/en active Pending
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107004413A (en) * | 2014-11-28 | 2017-08-01 | 微软技术许可有限责任公司 | Expanding digital personal assistant acts supplier |
CN105068782A (en) * | 2015-08-11 | 2015-11-18 | 胡达广 | Recognition system for voice instruction of robot |
CN105807261A (en) * | 2016-04-26 | 2016-07-27 | 深思考人工智能机器人科技(北京)有限公司 | Device and method for distinguishing position by listening to sound |
CN105807261B (en) * | 2016-04-26 | 2018-04-10 | 深思考人工智能机器人科技(北京)有限公司 | A kind of audition distinguishes the device and method of position |
CN106378786A (en) * | 2016-11-30 | 2017-02-08 | 北京百度网讯科技有限公司 | Robot based on artificial intelligence |
CN106378786B (en) * | 2016-11-30 | 2018-12-21 | 北京百度网讯科技有限公司 | Robot based on artificial intelligence |
CN109831709A (en) * | 2019-02-15 | 2019-05-31 | 杭州嘉楠耘智信息科技有限公司 | Sound source orientation method and device and computer readable storage medium |
CN110193834A (en) * | 2019-05-16 | 2019-09-03 | 深圳市百思智能科技有限公司 | A kind of intelligent tracking system based on speech recognition |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101382590A (en) | Sound source direction recognizing method, recognizing device and robot with the device | |
WO2018137704A1 (en) | Microphone array-based pick-up method and system | |
CN110415695A (en) | A kind of voice awakening method and electronic equipment | |
KR101902717B1 (en) | Voice wake-up method and device | |
CN102932212A (en) | Intelligent household control system based on multichannel interaction manner | |
CN107680594B (en) | Distributed intelligent voice acquisition and recognition system and acquisition and recognition method thereof | |
CN106358120A (en) | Audio play device with various regulation methods | |
CN105976814A (en) | Headset control method and device | |
AU1740801A (en) | Methods and apparatuses for signal analysis | |
CN110110616A (en) | A kind of electronic equipment and control method | |
CN105182763A (en) | Intelligent remote controller based on voice recognition and realization method thereof | |
CN101966849B (en) | Method, device and system for controlling vehicle-mounted equipment | |
TWI815343B (en) | Noise reduction processing method | |
CN111583898B (en) | Space environment multi-directional selective noise reduction system and method | |
CN202524486U (en) | Voice control television | |
CN105049802A (en) | Speech recognition law-enforcement recorder and recognition method thereof | |
CN203457266U (en) | Voice acquisition device and television system with voice acquisition function | |
CN105491336A (en) | Image identification module with low power consumption | |
CN206819731U (en) | A kind of novel voice identification equipment | |
CN106935246A (en) | A kind of voice acquisition methods and electronic equipment based on microphone array | |
CN103997381B (en) | The identification of examination hall cheating signal intelligent and evidence obtaining method of reducing | |
CN206758124U (en) | A kind of voice print identification device based on cochlea cepstrum coefficient under noise circumstance | |
CN208925781U (en) | A kind of intelligent sound electrically driven curtain | |
CN214954915U (en) | Sign language capturing and translating equipment device | |
CN112073866A (en) | Intelligent earphone device with computer vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Open date: 20090311 |