CN102110441A - Method for generating sound masking signal based on time reversal - Google Patents
Method for generating sound masking signal based on time reversal Download PDFInfo
- Publication number
- CN102110441A CN102110441A CN2010106171655A CN201010617165A CN102110441A CN 102110441 A CN102110441 A CN 102110441A CN 2010106171655 A CN2010106171655 A CN 2010106171655A CN 201010617165 A CN201010617165 A CN 201010617165A CN 102110441 A CN102110441 A CN 102110441A
- Authority
- CN
- China
- Prior art keywords
- signal
- sound source
- frame
- time reversal
- target sound
- 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
- Soundproofing, Sound Blocking, And Sound Damping (AREA)
Abstract
The invention relates to a method for generating a sound masking signal based on time reversal. In the method, a corresponding sound masking signal is acquired according to a target sound source signal and has a long-time amplitude spectrum similar to that of the target sound source signal and extremely low speech intelligibility. The method comprises the following steps of: picking up the target sound source signal by using a microphone or a microphone array and performing preprocessing to acquire a clean target sound source signal; performing framing based on a specific time window length according to the acquired target sound source signal; and performing time reversal on a time domain of each frame of signal according to y(t)=x(-t) to acquire the sound masking signal corresponding to a target sound source, wherein x(t) represents a frame of signal, and the y(t) represents an output signal of the frame. The time reversal sound masking signal has the long-time amplitude spectrum similar to that of the target sound source signal; the method has advantages in energy masking; and the time reversal sound masking signal is completely unintelligible or is partially intelligible and cannot become a new interference sound source.
Description
Technical field
The present invention relates to the production method of sound masking signal, particularly a kind of sound masking signal generating method based on time reversal.
Background technology
Studies show that voice are in the room people to be disturbed maximum voice signal, the speech intelligibilty of voice signal can reduce work efficiency.In some occasion, because speech intelligibilty is too high, individual's conversation privacy can not get ensureing, is badly in need of a kind of method that can reduce speech intelligibilty and protection individual conversation privacy.Sound masking is meant that the sound with a kind of nature or synthetic joins in the environment, comes the coverage goal source sound by auditory masking, reaches the method that reduces target sound source signal intelligibility.At present, the acoustics macking technique based on the sound masking signal is considered to one of requisite measure of improving the open office acoustic enviroment.Existing sound masking signal generating method, their masking signal source generally is a noise signal, such as: white noise, pink noise, air-conditioning noise and the artificial various noises that produce, but noise signal is not owing to have correlativity with the target sound source signal, it is very low to shelter efficient, this just need make the noise energy at acceptance point place much larger than the target sound source signal energy, could reduce the speech intelligibilty of target sound source.Problem is that strong excessively noise energy can increase the worried degree of sound, and the people can't be stood.If the sound masking signal is produced by the target sound source signal Processing, both have correlativity, can improve and shelter efficient.Therefore, need to seek a kind of more efficiently sound masking signal, this signal is obtained by the target sound source signal Processing, is better than the masking by noise signal aspect the efficient sheltering.
Summary of the invention
The objective of the invention is to, the present invention proposes a kind of sound masking signal generating method based on time reversal, and the sound masking signal is produced by the target sound source signal Processing, and both have correlativity, can improve and shelter efficient.
For achieving the above object, the invention provides a kind of sound masking signal generating method based on time reversal, this method obtains corresponding sound masking signal according to the target sound source signal, amplitude spectrum when this sound masking signal has with target sound source signal similar long, and speech intelligibilty is very low; These method concrete steps comprise:
Step 1): use microphone or microphone array to pick up the target sound source signal, obtain clean target sound source signal by pre-service;
Step 2): the target sound source signal that obtains according to described step 1) carries out the branch frame by special time window length, according to formula (1) each frame signal is carried out obtaining after time reversal of time domain the sound masking signal of target sound source;
y(t)＝x(-t) (1)
Wherein, x (t) represents a frame signal, and y (t) represents the output signal of this frame.
Pre-service in the described step 1) comprises: voice enhancing, noise reduction, auditory localization and sound feature identification.
Frame length is 150ms～500ms when carrying out the branch frame by special time window length described step 2).Described step 2) also comprises: level and smooth every frame signal joining day time reversal window function.Described time window function edge is decayed fast and the end points place is 0.
The invention has the advantages that, sound masking time reversal signal of the present invention and target sound source signal have similar amplitude spectrum when long, on energy is sheltered, preponderate, and sound masking time reversal signal can not be understood fully or part can be understood, can not become new interference sound source.The masking performance of sound masking time reversal signal is far above the masking by noise signal, at identical target sound source signal to the energy of sound masking signal than under the situation, can reduce speech intelligibilty greatly, protection conversation privacy.In addition, sound masking time reversal signal generating method, signal processing is simple, is fit to handle in real time.
Description of drawings
Fig. 1 is a kind of sound masking signal generating method process flow diagram based on time reversal of the present invention;
Fig. 2 is the oscillogram after the branch frame is carried out in target sound source signal process of the present invention pre-service;
Fig. 3 is based on the oscillogram after waveform among Fig. 2 carries out time reversal;
Fig. 4 is based on waveform among Fig. 3 and adds the sound masking signal waveforms of Tukey window function after level and smooth;
Fig. 5 be target sound source signal among Fig. 2 and the stack of the sound masking signal among Fig. 4 obtain shelter oscillogram.
Embodiment
Below in conjunction with drawings and Examples the present invention is further specified.
Amplitude spectrum when this sound masking signal has with target sound source signal similar long, and speech intelligibilty is very low.The sound masking signal that the inventive method produces is being sheltered aspect the efficient; be much better than existing masking by noise signal; show that identical target sound source signal harmony masking signal energy is than under TMR (Target-to-Masker ratio) situation; make the speech intelligibilty of target sound source signal lower, effectively protection conversation privacy.
A kind of sound masking signal generating method process flow diagram of the present invention based on time reversal, as shown in Figure 1.Technical scheme of the present invention comprises the steps:
Step 1): single microphone, a plurality of microphone or microphone array pick up the target sound source signal, carry out pre-service, and preprocessing process comprises voice enhancing, noise reduction etc.
Step 2): pretreated clean target sound source signal is carried out the branch frame by special time window length, each frame signal is carried out the time reversal of time domain, to each the frame signal joining day window function after time reversal, obtain the sound masking signal of target sound source between the frame of level and smooth front and back.The sound masking signal is reset by speaker system, and the target sound source signal is produced masking action, reduces the speech intelligibilty of target sound source signal, the conversation privacy of protection target sound source
Below each step of the present invention is described in further detail in conjunction with Fig. 2, Fig. 3, Fig. 4 and Fig. 5:
In the described step 1, be implemented as follows:
Pick up the target sound source signal, realize by single microphone, a plurality of microphone or microphone array, because the existence of ground unrest and the difference of target sound source and microphone position, the signal that picks up may comprise noise and other signals, obtain clean target sound source signal by pre-service, pretreated process can comprise voice enhancing, noise reduction, auditory localization, sound source feature identification etc.
In the described step 2, be implemented as follows:
Target sound source signal of the present invention carries out oscillogram behind the branch frame through pre-service, and as shown in Figure 2, pretreated signal carries out the branch frame by specific time window length, the recommended range of frame be 150ms between the 500ms, be exemplified as 200ms among the figure.Then as shown in Figure 3, the signal behind minute frame is carried out time reversal handle, each frame signal is reversed on time domain, the signal after obtaining reversing.At this moment, this signal is discontinuous at frame and frame junction, is undertaken smoothly by window function, and level and smooth back signal as shown in Figure 4.
Time reversal signal, along with frame length changes, speech intelligibilty can change, and is in particular in frame length less than 50ms, and time reversal, signal almost can be understood fully, along with frame length increases, speech intelligibilty descends, and when frame length was 130ms, the chances are 50% can understand,, can not understand fully near 200ms up to frame length.Amplitude spectrum is similar when wishing design to the target sound source Chief Signal Boatswain in this invention, but the masking signal that can not understand fully or partly can understand, and so frame length is recommended as more than the 150ms, if frame length is too short, itself can understand the sound masking signal, can't shelter.Frame length is greater than more than the 200ms in theory, and time reversal, signal can not be understood fully, but the requirement of handling in real time in considering to use, and the frame length scope is recommended in below the 500ms.
Signal after time reversal such as Fig. 3 are discontinuous between frame and the frame, discontinuously may produce noise, increase the worried degree of masking signal time reversal, and service time, window function carried out smoothly here.For the selection of window function, requiring the fast and end points place of window function edge decay is 0, most of shape information of retention time counter-rotating sound masking signal as far as possible, and Fig. 4 is for through the level and smooth back of window function frame endpoint value being 0 signal waveforms.
The sound masking signal of target sound source as shown in Figure 4 in the legend, reset by speaker system, speaker system can be single loudspeaker, a plurality of loudspeaker or loudspeaker array, according to different system space project organizations and the desired masking effect that reaches, adjust the input signal amplitude of loudspeaker, realize speech intelligibilty control the target sound source signal.Generally speaking, acceptance point place target sound source signal to the energy of sound masking time reversal signal than at-5dB between the 0dB, speech intelligibilty has decline to a certain degree, if wish lower speech intelligibilty, can suitably improve the energy of sound masking time reversal signal.
The specific embodiment of the invention is as follows:
The present embodiment signal processing carries out emulation in MATLAB software, result is by sense of hearing subjective experiment evaluation speech intelligibilty.Suppose to obtain clean target sound source signal, as shown in Figure 2, then signal among Fig. 2 is carried out the branch frame processing that 200ms is a frame, dotted line is for dividing frame boundaries as shown in Figure 2, then every frame signal being carried out time reversal handles, as shown in Figure 3, at this moment, be discontinuous between frame and the frame, discontinuous in order to eliminate, it is level and smooth to add the Tukey window function, window function MATLAB expression formula is " tukeywin (L, 0.2) ", and wherein L is a sampling number in the frame, obtain the sound masking time reversal signal of this target sound source signal at last, as shown in Figure 4.Regulate the energy ratio of target sound source signal graph 2 harmony masking signal Fig. 4, obtain after the stack exporting test signal as shown in Figure 5, wherein, the result when Fig. 5 is 0dB for the energy ratio.The speech intelligibilty error rate of test pattern 5 outputs, use noise signal simultaneously as a reference, experimental result is as shown in table 1, show sound masking time reversal signal at identical energy than under the condition, compare the masking by noise signal, has better reduction speech intelligibilty ability, such as energy than for-10dB the time, sound masking time reversal signal covers over the object behind the sound-source signal, the intelligibility error rate is 97%, do not have speech intelligibilty, and the masking by noise signal covers over the object behind the sound-source signal, speech intelligibilty is not forfeiture almost.
The error rate of table 1 experimental result-speech intelligibilty
| Energy compares TMR | -15dB | -10dB | -5dB | 0dB |
| Sound masking time reversal signal | 96％ | 97％ | 31％ | 8％ |
| The masking by noise signal | 12.5％ | 0.13％ | 0.2％ | 0.1％ |
In the present embodiment, though adopted 200ms as frame length and adopt Tukey as time window, this only is that to method provided by the present invention one illustrates, and above embodiment is only unrestricted in order to technical scheme of the present invention to be described.Although the present invention is had been described in detail with reference to embodiment, those of ordinary skill in the art is to be understood that, technical scheme of the present invention is made amendment or is equal to replacement, do not break away from the spirit and scope of technical solution of the present invention, it all should be encompassed in the middle of the claim scope of the present invention.
Claims (6)
1. sound masking signal generating method based on time reversal, this method obtains corresponding sound masking signal according to the target sound source signal, amplitude spectrum when this sound masking signal has with target sound source signal similar long, and speech intelligibilty is very low; These method concrete steps comprise:
Step 1): use microphone or microphone array to pick up the target sound source signal, obtain clean target sound source signal by pre-service;
Step 2): the target sound source signal that obtains according to described step 1) carries out the branch frame by special time window length, each frame signal is carried out the corresponding sound masking signal that obtains target sound source time reversal of time domain according to formula (1);
y(t)＝x(-t) (1)
Wherein, x (t) represents a frame signal, and y (t) represents the output signal of this frame.
2. the sound masking signal generating method based on time reversal according to claim 1 is characterized in that the pre-service in the described step 1) comprises: voice enhancing, noise reduction, auditory localization and sound feature identification.
3. the sound masking signal generating method based on time reversal according to claim 1 is characterized in that described step 2) in when carrying out the branch frame by special time window length frame length be 150ms～500ms.
4. the sound masking signal generating method based on time reversal according to claim 1 is characterized in that described step 2) also comprise: level and smooth to every frame signal joining day time reversal window function.
5. the sound masking signal generating method based on time reversal according to claim 4 is characterized in that, described time window function edge is decayed fast and the end points place is 0.
6. according to claim 4 or 5 described sound masking signal generating methods, it is characterized in that described time window function is the Tukey window function based on time reversal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106171655A CN102110441A (en) | 2010-12-22 | 2010-12-22 | Method for generating sound masking signal based on time reversal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2010106171655A CN102110441A (en) | 2010-12-22 | 2010-12-22 | Method for generating sound masking signal based on time reversal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102110441A true CN102110441A (en) | 2011-06-29 |
Family
ID=44174574
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2010106171655A Pending CN102110441A (en) | 2010-12-22 | 2010-12-22 | Method for generating sound masking signal based on time reversal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102110441A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102522080A (en) * | 2011-12-08 | 2012-06-27 | 中国科学院声学研究所 | Random interference sound signal generating system and method for protecting language privacy |
| CN103886858A (en) * | 2014-03-11 | 2014-06-25 | 中国科学院信息工程研究所 | Sound masking signal generating method and system |
| CN104347076A (en) * | 2013-08-09 | 2015-02-11 | 中国电信股份有限公司 | Network audio packet loss concealment method and device |
| CN104575486A (en) * | 2014-12-25 | 2015-04-29 | 中国科学院信息工程研究所 | Sound leakage protection method and system based on sound masking principle |
| CN105185370A (en) * | 2015-08-10 | 2015-12-23 | 电子科技大学 | Sound masking door |
| CN105493177A (en) * | 2013-08-22 | 2016-04-13 | 微软技术许可有限责任公司 | Preserving privacy of a conversation from surrounding environment |
| CN109862472A (en) * | 2019-02-21 | 2019-06-07 | 中科上声（苏州）电子有限公司 | A kind of car privacy call method and system |
| CN110007276A (en) * | 2019-04-18 | 2019-07-12 | 太原理工大学 | A kind of sound localization method and system |
| CN112104439A (en) * | 2020-11-23 | 2020-12-18 | 北京中超伟业信息安全技术股份有限公司 | Self-adaptive recording interference method and system |
| CN113497849A (en) * | 2020-03-20 | 2021-10-12 | 华为技术有限公司 | Sound masking method and device and terminal equipment |
| CN114337908A (en) * | 2022-01-05 | 2022-04-12 | 中国科学院声学研究所 | Method and device for generating interference signal of target voice signal |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040019479A1 (en) * | 2002-07-24 | 2004-01-29 | Hillis W. Daniel | Method and system for masking speech |
| CN1514431A (en) * | 2003-08-08 | 2004-07-21 | 中国科学院声学研究所 | Non linear spectrum reduction and missing component estimation method |
| WO2006076217A2 (en) * | 2005-01-10 | 2006-07-20 | Herman Miller, Inc. | Method and apparatus of overlapping and summing speech for an output that disrupts speech |
| CN101218768A (en) * | 2005-10-07 | 2008-07-09 | 株式会社Ntt都科摩 | Modulation device, modulation method, demodulation device, and demodulation method |
| US20080235008A1 (en) * | 2007-03-22 | 2008-09-25 | Yamaha Corporation | Sound Masking System and Masking Sound Generation Method |
-
2010
- 2010-12-22 CN CN2010106171655A patent/CN102110441A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040019479A1 (en) * | 2002-07-24 | 2004-01-29 | Hillis W. Daniel | Method and system for masking speech |
| CN1514431A (en) * | 2003-08-08 | 2004-07-21 | 中国科学院声学研究所 | Non linear spectrum reduction and missing component estimation method |
| WO2006076217A2 (en) * | 2005-01-10 | 2006-07-20 | Herman Miller, Inc. | Method and apparatus of overlapping and summing speech for an output that disrupts speech |
| CN101218768A (en) * | 2005-10-07 | 2008-07-09 | 株式会社Ntt都科摩 | Modulation device, modulation method, demodulation device, and demodulation method |
| US20080235008A1 (en) * | 2007-03-22 | 2008-09-25 | Yamaha Corporation | Sound Masking System and Masking Sound Generation Method |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102522080B (en) * | 2011-12-08 | 2013-12-11 | 中国科学院声学研究所 | Random interference sound signal generating system and method for protecting language privacy |
| CN102522080A (en) * | 2011-12-08 | 2012-06-27 | 中国科学院声学研究所 | Random interference sound signal generating system and method for protecting language privacy |
| CN104347076B (en) * | 2013-08-09 | 2017-07-14 | 中国电信股份有限公司 | Network audio packet loss covering method and device |
| CN104347076A (en) * | 2013-08-09 | 2015-02-11 | 中国电信股份有限公司 | Network audio packet loss concealment method and device |
| CN105493177B (en) * | 2013-08-22 | 2020-04-07 | 微软技术许可有限责任公司 | System and computer-readable storage medium for audio processing |
| CN105493177A (en) * | 2013-08-22 | 2016-04-13 | 微软技术许可有限责任公司 | Preserving privacy of a conversation from surrounding environment |
| CN103886858A (en) * | 2014-03-11 | 2014-06-25 | 中国科学院信息工程研究所 | Sound masking signal generating method and system |
| CN103886858B (en) * | 2014-03-11 | 2016-10-05 | 中国科学院信息工程研究所 | A kind of sound masking signal generating method and system |
| CN104575486A (en) * | 2014-12-25 | 2015-04-29 | 中国科学院信息工程研究所 | Sound leakage protection method and system based on sound masking principle |
| CN104575486B (en) * | 2014-12-25 | 2019-04-02 | 中国科学院信息工程研究所 | Sound leakage protection method and system based on the principle of acoustic masking |
| CN105185370B (en) * | 2015-08-10 | 2019-02-12 | 电子科技大学 | A kind of sound masking door |
| CN105185370A (en) * | 2015-08-10 | 2015-12-23 | 电子科技大学 | Sound masking door |
| CN109862472A (en) * | 2019-02-21 | 2019-06-07 | 中科上声（苏州）电子有限公司 | A kind of car privacy call method and system |
| CN110007276A (en) * | 2019-04-18 | 2019-07-12 | 太原理工大学 | A kind of sound localization method and system |
| CN110007276B (en) * | 2019-04-18 | 2021-01-12 | 太原理工大学 | Sound source positioning method and system |
| CN113497849A (en) * | 2020-03-20 | 2021-10-12 | 华为技术有限公司 | Sound masking method and device and terminal equipment |
| CN112104439A (en) * | 2020-11-23 | 2020-12-18 | 北京中超伟业信息安全技术股份有限公司 | Self-adaptive recording interference method and system |
| CN114337908A (en) * | 2022-01-05 | 2022-04-12 | 中国科学院声学研究所 | Method and device for generating interference signal of target voice signal |
| CN114337908B (en) * | 2022-01-05 | 2024-04-12 | 中国科学院声学研究所 | Method and device for generating interference signal of target voice signal |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102110441A (en) | Method for generating sound masking signal based on time reversal | |
| CN106251877B (en) | Voice Sounnd source direction estimation method and device | |
| CN103873977B (en) | Recording system and its implementation based on multi-microphone array beam forming | |
| JP5675848B2 (en) | Adaptive noise suppression by level cue | |
| CN102938254B (en) | Voice signal enhancement system and method | |
| Rui et al. | Time delay estimation in the presence of correlated noise and reverberation | |
| CN105869651B (en) | Binary channels Wave beam forming sound enhancement method based on noise mixing coherence | |
| CN105741849A (en) | Voice enhancement method for fusing phase estimation and human ear hearing characteristics in digital hearing aid | |
| CN101996630A (en) | Automatic sound recognition based on binary time frequency unit | |
| CN101278337A (en) | Robust separation of speech signals in a noisy environment | |
| EP3245795B1 (en) | Reverberation suppression using multiple beamformers | |
| CN103871421A (en) | Self-adaptive denoising method and system based on sub-band noise analysis | |
| CN106297817B (en) | A kind of sound enhancement method based on binaural information | |
| CN109218882A (en) | The ambient sound monitor method and earphone of earphone | |
| CN106970356A (en) | Auditory localization tracking under a kind of complex environment | |
| CN109874096A (en) | A kind of ears microphone hearing aid noise reduction algorithm based on intelligent terminal selection output | |
| CN104661152A (en) | Spatial filterbank for hearing system | |
| CN108986832A (en) | Ears speech dereverberation method and device based on voice probability of occurrence and consistency | |
| CN102522080B (en) | Random interference sound signal generating system and method for protecting language privacy | |
| Liu | Sound source seperation with distributed microphone arrays in the presence of clocks synchronization errors | |
| CN108235208A (en) | For running the method for hearing aid apparatus | |
| MX2022001162A (en) | Acoustic echo cancellation control for distributed audio devices. | |
| Henry et al. | Noise reduction in cochlear implant signal processing: A review and recent developments | |
| Wittkop et al. | Speech processing for hearing aids: Noise reduction motivated by models of binaural interaction | |
| Ganguly et al. | Real-time smartphone application for improving spatial awareness of hearing assistive devices |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110629 |