CN105096946B - Awakening device and method based on voice activation detection - Google Patents
Awakening device and method based on voice activation detection Download PDFInfo
- Publication number
- CN105096946B CN105096946B CN201410193199.4A CN201410193199A CN105096946B CN 105096946 B CN105096946 B CN 105096946B CN 201410193199 A CN201410193199 A CN 201410193199A CN 105096946 B CN105096946 B CN 105096946B
- Authority
- CN
- China
- Prior art keywords
- signal
- voice
- wake
- fundamental frequency
- unit
- 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
Images
Landscapes
- Emergency Alarm Devices (AREA)
- Measurement Of Velocity Or Position Using Acoustic Or Ultrasonic Waves (AREA)
Abstract
The invention relates to the technical field of electronics, in particular to a wake-up device and a wake-up method. The awakening device based on voice activation detection comprises a microphone, a voice detection module and a control module, wherein the microphone is used for sensing a sound signal and converting the sound signal into an electric signal; the first signal extraction unit is connected with the output end of the microphone and used for extracting a fundamental frequency signal in the output signal of the microphone; the comparison unit is connected with the first signal extraction unit and compares the fundamental frequency signal with a reference fundamental frequency range representing the characteristics of the human voice to generate a first comparison result; and the controller is connected with the comparison unit and generates a wake-up signal under the action of the output signal of the comparison unit. The invention can not only identify whether the voice signal meets the voice characteristics, but also identify the distance between the speaker, so as to avoid the irrelevant voice signal from interfering the normal voice activation detection and effectively wake up the voice.
Description
Technical Field
The invention relates to the technical field of electronics, in particular to a wake-up device and a wake-up method.
Background
Along with the continuous integration and development of electronic devices, electronic devices with voice recognition technology are becoming more popular, which can control electronic devices to execute related functions through voice commands when users are inconvenient to operate the electronic devices, wherein the function of waking up the electronic devices through the voice commands is gaining wide attention, in the voice recognition technology of the prior art, the wake-up distance is not controllable, and misjudgment is often generated for long-distance noisy human voices, for example, the voice at 2 meters or 5 meters far away is responded, and the normal use of the users is disturbed; and turning on the voice wake-up system too frequently causes excessive system power consumption.
Disclosure of Invention
The invention aims to provide a wake-up device based on voice activation detection, which solves the technical problems;
the present invention also aims to provide a wake-up method based on voice activation detection, which solves the above technical problems.
The technical problem solved by the invention can be realized by adopting the following technical scheme:
wake-up device based on voice activation detection, comprising,
a microphone for sensing the sound signal and converting it into electric signal;
the first signal extraction unit is connected with the output end of the microphone and used for extracting a fundamental frequency signal in the output signal of the microphone;
the comparison unit is connected with the first signal extraction unit and is used for comparing the fundamental frequency signal with a reference fundamental frequency range representing the human voice characteristics to generate a first comparison result;
and the controller is connected with the comparison unit and generates a wake-up signal under the action of the output signal of the comparison unit.
Preferably, the microphone further comprises a second signal extraction unit, connected to the output end of the microphone, for extracting a higher harmonic signal from the output signal of the microphone;
the comparison unit is further connected with the output end of the second signal extraction unit and used for comparing the higher harmonic signal with a set limit value to generate a second comparison result.
Preferably, the microphone further comprises a frequency domain transforming unit, an output end of the microphone is connected to the frequency domain transforming unit, and is configured to perform frequency domain transformation on the output signal of the microphone, and an output end of the frequency domain transforming unit is connected to the first signal extracting unit and the second signal extracting unit, respectively.
Preferably, the first signal extraction unit employs a band-pass filter, and the second signal extraction unit employs a high-pass filter.
Preferably, the frequency domain transforming unit adopts a fourier transformer.
Preferably, the device further comprises a switching device connected with the controller and switching between a sleep mode and an operating mode under the action of the wake-up signal.
Preferably, the device further comprises a storage unit, which stores the reference fundamental frequency range and the setting limit value.
The invention also provides a voice activation detection-based awakening method, which comprises the following specific steps:
step 1, sensing an external sound signal and converting the external sound signal into an electric signal;
step 2, respectively extracting a fundamental frequency signal and a higher harmonic signal in the electric signal;
and 4, comparing the higher harmonic signal with a set limit value on the basis of judging that the higher harmonic signal is the voice, indicating that the sound source distance is far and not responding when the higher harmonic signal is smaller than the set limit value, and indicating that the sound source distance is close and generating a wake-up signal when the higher harmonic signal is larger than the set limit value.
Preferably, a step of frequency domain transformation of the electrical signal is further included between step 1 and step 2.
Preferably, the frequency domain transform employs a fourier transform.
Has the advantages that: by adopting the technical scheme, the voice activation detection method and the voice activation detection device can not only identify whether the voice signal meets the voice characteristics, but also identify the distance between the speaker, so that the irrelevant voice signal is prevented from interfering the normal voice activation detection, and the voice awakening can be effectively carried out.
Drawings
FIG. 1 is a schematic circuit diagram of the present invention;
FIG. 2 is another circuit schematic of the present invention;
FIG. 3 is a flowchart of a wake-up method according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
Referring to fig. 1, a wake-up unit based on voice activity detection includes, among other things,
a microphone 1 for sensing the sound signal and converting it into an electric signal;
a first signal extraction unit 21, connected to the output end of the microphone 1, for extracting the fundamental frequency signal from the output signal of the microphone 1;
the comparison unit 3 is connected with the first signal extraction unit 21, and compares the fundamental frequency signal with a reference fundamental frequency range representing the characteristics of the human voice to generate a first comparison result;
and the controller 4 is connected with the comparison unit 3 and generates a wake-up signal under the action of the output signal of the comparison unit 3.
The invention extracts the fundamental frequency signals in the sound signals by utilizing the fact that the sound emitted by a person has a specific fundamental frequency range, the transmission attenuation of the fundamental frequency is small, the sound can be transmitted for a relatively long distance, and the fundamental frequency signals are compared, and because of the difference in the physiological aspects of sounding organs, the fundamental frequency ranges of males and females are different, the fundamental frequency range of males is 50-250 Hz, the fundamental frequency range of females is 120-500 Hz, and the reference fundamental frequency range can be stored in a storage unit or can be generated by a reference fundamental frequency signal generator.
Because the fundamental frequency can identify who speaks, furthermore, a specific human fundamental frequency signal can be simulated through a simulation signal generator to be used as a reference fundamental frequency signal, and when the fundamental frequency signal in the voice signal is detected to be matched with the reference fundamental frequency signal, subsequent processing is carried out.
As a preferred embodiment of the present invention, the present invention further comprises a second signal extraction unit 22, connected to the output terminal of the microphone 1, for extracting a higher harmonic signal from the output signal of the microphone 1;
the comparing unit 3 is further connected to the output terminal of the second signal extracting unit 22, and is configured to compare the higher harmonic signal with a set limit value, and generate a second comparison result.
High frequency of sound is easy to absorb and attenuate in the transmission process, and the attenuation of a signal of certain frequency of the sound in the air has a qualitative rule; by analyzing the harmonic components in the signal acquired by the microphone 1, when the high frequency components are attenuated to a value lower than a predetermined limit value, it is considered that the sound source distance is long and no judgment is made. The invention can not only identify whether the voice signal meets the voice characteristics, but also identify the distance between the speakers so as to determine the effective voice, so as to avoid the irrelevant voice signal from interfering the normal voice activation detection and effectively carry out voice awakening. The invention preferably uses the attenuation ratio after propagation for a distance of 1 meter or 1.5 meters as a set limit for a signal of a certain frequency.
As a preferred embodiment of the present invention, referring to fig. 2, the extraction of the fundamental frequency signal may be extracted from a time domain signal or a frequency domain signal, and may include a frequency domain transforming unit 6, where an output end of the microphone 1 is connected to the frequency domain transforming unit 6, and is used for performing frequency domain transformation on the output signal of the microphone 1, and output ends of the frequency domain transforming unit 6 are respectively connected to the first signal extracting unit 21 and the second signal extracting unit 22; the frequency domain transforming unit 6 may be implemented using a fourier transformer.
As a preferred embodiment of the present invention, the first signal extraction unit 21 may employ a band pass filter. The second signal extraction unit may employ a high pass filter. It can also be combined with the prior art extraction method of fundamental frequency or higher harmonic signals.
As a preferred embodiment of the present invention, a switching device 7 is further included, which is connected to the controller 4 and switches between the sleep mode and the operating mode under the action of the wake-up signal.
As a preferred embodiment of the present invention, a memory unit 5 is included, which stores the reference fundamental frequency range and the setting limit value.
Referring to fig. 3, the present invention further provides a wake-up method based on voice activation detection, wherein the specific steps are as follows:
step 1, sensing an external sound signal and converting the external sound signal into an electric signal;
step 2, respectively extracting a fundamental frequency signal and a higher harmonic signal in the electric signal;
and 4, comparing the higher harmonic signal with a set limit value on the basis of judging the higher harmonic signal as the voice, indicating that the sound source distance is far and no response is made when the higher harmonic signal is smaller than the set limit value, and indicating that the sound source distance is close and generating a wake-up signal when the higher harmonic signal is larger than the set limit value.
As a preferred embodiment of the present invention, a step of frequency domain transforming the electrical signal is further included between step 1 and step 2.
As a preferred embodiment of the present invention, the frequency domain transform employs a fourier transform.
The invention determines the effective voice by identifying whether the frequency spectrum meets the voice characteristics and identifying the distance between the speakers, avoids irrelevant voice signals from interfering the normal voice activation detection and can effectively carry out voice awakening.
While the invention has been described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.
Claims (9)
1. Wake-up device based on voice activity detection, comprising,
a microphone for sensing the sound signal and converting it into electric signal;
the first signal extraction unit is connected with the output end of the microphone and used for extracting a fundamental frequency signal in the output signal of the microphone;
the comparison unit is connected with the first signal extraction unit and is used for comparing the fundamental frequency signal with a reference fundamental frequency range representing the human voice characteristics to generate a first comparison result;
the controller is connected with the comparison unit and generates a wake-up signal under the action of an output signal of the comparison unit;
the device also comprises a reference base frequency signal generator which is used for generating a reference base frequency signal simulating a specific person;
the second signal extraction unit is connected with the output end of the microphone and used for extracting a higher harmonic signal in the output signal of the microphone;
the comparison unit is also connected with the output end of the second signal extraction unit and used for comparing the higher harmonic signal with a set limit value to generate a second comparison result;
and when the first comparison result shows that the fundamental frequency signal is located in the reference fundamental frequency range, the comparison unit compares the higher harmonic signal with the set limit value to generate the second comparison result.
2. The wake-up apparatus based on voice activation detection as claimed in claim 1, further comprising a frequency domain transforming unit, wherein an output of the microphone is connected to the frequency domain transforming unit for performing frequency domain transformation on the output signal of the microphone, and an output of the frequency domain transforming unit is connected to the first signal extracting unit and the second signal extracting unit, respectively.
3. The voice activation detection based wake-up apparatus according to claim 1, wherein the first signal extraction unit employs a band-pass filter, and the second signal extraction unit employs a high-pass filter.
4. The voice activation detection based wake-up apparatus according to claim 2, wherein the frequency domain transforming unit employs a fourier transformer.
5. A wake-up unit based on voice activity detection as claimed in claim 1, further comprising switching means connected to the controller for switching between a sleep mode and an operational mode under the influence of the wake-up signal.
6. The device according to claim 1, comprising a memory unit storing the reference base frequency range and the setting limit.
7. The wake-up method based on voice activation detection is applied to the wake-up device based on voice activation detection in any one of claims 1 to 6, and comprises the following specific steps:
step 1, sensing an external sound signal and converting the external sound signal into an electric signal;
step 2, respectively extracting a fundamental frequency signal and a higher harmonic signal in the electric signal;
step 3, comparing the fundamental frequency signal with a reference fundamental frequency range representing the characteristics of the voice, and judging the voice when the fundamental frequency signal is positioned in the reference fundamental frequency range;
and 4, comparing the higher harmonic signal with a set limit value on the basis of judging that the higher harmonic signal is the voice, indicating that the sound source distance is far and not responding when the higher harmonic signal is smaller than the set limit value, and indicating that the sound source distance is close and generating a wake-up signal when the higher harmonic signal is larger than the set limit value.
8. The voice activation detection based wake-up method according to claim 7, further comprising a step of frequency domain transforming the electrical signal between step 1 and step 2.
9. The voice activity detection based wake-up method according to claim 8, wherein the frequency domain transform employs a fourier transform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410193199.4A CN105096946B (en) | 2014-05-08 | 2014-05-08 | Awakening device and method based on voice activation detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410193199.4A CN105096946B (en) | 2014-05-08 | 2014-05-08 | Awakening device and method based on voice activation detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105096946A CN105096946A (en) | 2015-11-25 |
| CN105096946B true CN105096946B (en) | 2020-09-29 |
Family
ID=54577232
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410193199.4A Active CN105096946B (en) | 2014-05-08 | 2014-05-08 | Awakening device and method based on voice activation detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105096946B (en) |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10079027B2 (en) * | 2016-06-03 | 2018-09-18 | Nxp B.V. | Sound signal detector |
| CN106296890B (en) * | 2016-07-22 | 2019-06-04 | 北京小米移动软件有限公司 | Unlocking method, device and the mobile terminal of mobile terminal |
| CN106157950A (en) * | 2016-09-29 | 2016-11-23 | 合肥华凌股份有限公司 | Speech control system and awakening method, Rouser and household electrical appliances, coprocessor |
| CN107564520A (en) * | 2017-09-01 | 2018-01-09 | 联想(北京)有限公司 | A kind of control method and electronic equipment |
| CN108377605A (en) * | 2018-04-03 | 2018-08-07 | 海盐瑞科光电科技有限公司 | A kind of bathroom voice control intelligent switch |
| CN108375923A (en) * | 2018-04-03 | 2018-08-07 | 海盐瑞科光电科技有限公司 | A kind of bathroom intelligent control switch |
| CN110648660A (en) * | 2018-06-27 | 2020-01-03 | 深圳联友科技有限公司 | Voice activation method of BS (base station) end |
| CN110097875B (en) * | 2019-06-03 | 2022-09-02 | 清华大学 | Microphone signal based voice interaction wake-up electronic device, method, and medium |
| CN111343410A (en) * | 2020-02-14 | 2020-06-26 | 北京字节跳动网络技术有限公司 | Mute prompt method and device, electronic equipment and storage medium |
| CN112995821B (en) * | 2021-03-01 | 2023-01-31 | 歌尔微电子股份有限公司 | Microphone control method, microphone system and wearable device |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6397186B1 (en) * | 1999-12-22 | 2002-05-28 | Ambush Interactive, Inc. | Hands-free, voice-operated remote control transmitter |
| JP4854533B2 (en) * | 2007-01-30 | 2012-01-18 | 富士通株式会社 | Acoustic judgment method, acoustic judgment device, and computer program |
| CN103052001B (en) * | 2011-10-17 | 2015-06-24 | 联想(北京)有限公司 | Intelligent device and control method thereof |
| CN203504751U (en) * | 2013-08-16 | 2014-03-26 | 上海耐普微电子有限公司 | Microphone with function of voice awakening and mobile terminal |
| CN203882609U (en) * | 2014-05-08 | 2014-10-15 | 钰太芯微电子科技（上海）有限公司 | Awakening apparatus based on voice activation detection |
-
2014
- 2014-05-08 CN CN201410193199.4A patent/CN105096946B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105096946A (en) | 2015-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105096946B (en) | Awakening device and method based on voice activation detection | |
| CN203882609U (en) | Awakening apparatus based on voice activation detection | |
| CN109087669B (en) | Audio similarity detection method and device, storage medium and computer equipment | |
| US9570072B2 (en) | System and method for noise reduction in processing speech signals by targeting speech and disregarding noise | |
| CN107477793A (en) | A kind of air purifier, control system of air purifier and method | |
| CN109920419B (en) | Voice control method and device, electronic equipment and computer readable medium | |
| WO2019127112A1 (en) | Voice interaction method and device and intelligent terminal | |
| WO2019233228A1 (en) | Electronic device and device control method | |
| EP3065043A1 (en) | Mobile device | |
| CN113949956B (en) | Noise reduction processing method and device, electronic equipment, earphone and storage medium | |
| US11250854B2 (en) | Method and apparatus for voice interaction, device and computer-readable storage medium | |
| Lezzoum et al. | Voice activity detection system for smart earphones | |
| JP6294747B2 (en) | Notification sound sensing device, notification sound sensing method and program | |
| CN112581960A (en) | Voice wake-up method and device, electronic equipment and readable storage medium | |
| EP3229487A1 (en) | Approach for detecting alert signals in changing environments | |
| CN115314804A (en) | Wearing detection method, wearable device and storage medium | |
| WO2017177629A1 (en) | Far-talking voice recognition method and device | |
| CN104282303A (en) | Method for conducting voice recognition by voiceprint recognition and electronic device thereof | |
| CN111402910B (en) | Method and equipment for eliminating echo | |
| CN103730117A (en) | Self-adaptation intelligent voice device and method | |
| CN112786047B (en) | Voice processing method, device, equipment, storage medium and intelligent sound box | |
| CN113808566B (en) | Vibration noise processing method and device, electronic equipment and storage medium | |
| CN114302286A (en) | Method, device and equipment for reducing noise of call voice and storage medium | |
| CN114429766A (en) | Method, device and equipment for adjusting playing volume and storage medium | |
| CN110839196B (en) | Electronic equipment and playing control method thereof |
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 |