CN117746880A - Intelligent noise reduction method and system for teleconference communication - Google Patents
Intelligent noise reduction method and system for teleconference communication Download PDFInfo
- Publication number
- CN117746880A CN117746880A CN202410185685.5A CN202410185685A CN117746880A CN 117746880 A CN117746880 A CN 117746880A CN 202410185685 A CN202410185685 A CN 202410185685A CN 117746880 A CN117746880 A CN 117746880A
- Authority
- CN
- China
- Prior art keywords
- noise
- signal
- audio
- conference
- conference room
- 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
- 230000009467 reduction Effects 0.000 title claims abstract description 107
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000004891 communication Methods 0.000 title claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 51
- 230000005236 sound signal Effects 0.000 claims abstract description 40
- 238000007781 pre-processing Methods 0.000 claims abstract description 27
- 230000008030 elimination Effects 0.000 claims abstract description 10
- 238000003379 elimination reaction Methods 0.000 claims abstract description 10
- 230000002708 enhancing effect Effects 0.000 claims abstract description 5
- 230000007613 environmental effect Effects 0.000 claims description 31
- 238000004458 analytical method Methods 0.000 claims description 14
- 238000009413 insulation Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 11
- 238000005516 engineering process Methods 0.000 claims description 9
- 238000010219 correlation analysis Methods 0.000 claims description 4
- 230000000694 effects Effects 0.000 abstract description 9
- 230000001629 suppression Effects 0.000 abstract 1
- 238000004422 calculation algorithm Methods 0.000 description 7
- 238000001914 filtration Methods 0.000 description 7
- 238000005259 measurement Methods 0.000 description 5
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000004378 air conditioning Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 239000012774 insulation material Substances 0.000 description 2
- 230000003993 interaction Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003472 neutralizing effect Effects 0.000 description 1
- 230000001681 protective effect Effects 0.000 description 1
- 230000008054 signal transmission Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Abstract
The invention relates to the technical field of telecommunications, and provides an intelligent noise reduction method and system for teleconference communication, comprising the following steps: the microphone array is used for collecting audio signals to obtain environment signals and audio preprocessing signals, and meanwhile, voice information of participants is obtained and voice recognition is carried out. Based on the voice signal, the voice enhancement module is used for enhancing the audio preprocessing signal to obtain an audio primary processing signal; the environment signals and the audio primary processing signals are combined to perform multi-source noise elimination, the audio processing signals are obtained and played, the technical problems that in a multi-terminal connection state of a teleconference, parameter adjustment for noise suppression is relatively complex and cannot meet real-time communication requirements are solved, the interference of various noises inside and outside a conference room is effectively reduced, the voice quality is improved, in the multi-terminal connection state of the teleconference, multi-source voice with noise is processed in real time, conference participants can hear the voices of all parties at different connection ends more clearly, and therefore the efficiency and quality technical effects of the conference are improved.
Description
Technical Field
The invention relates to the technical field of telecommunication, in particular to an intelligent noise reduction method and system for teleconference communication.
Background
In applications such as mobile communications, headphones, conference systems, voice interactions, etc., the voice noise reduction technique improves voice quality and intelligibility by removing noise and extracting clean voice, and has great value. The noise reduction method is mainly based on a signal processing technology, and suppresses noise by analyzing and processing an audio signal.
However, various parameters, such as a filter coefficient, a threshold value, etc., need to be adjusted in the process of processing to suppress noise, so as to optimize the noise reduction effect, and the adjustment of various parameters is very difficult, needs to have certain expertise and skills, and needs different parameter settings in different noise environments and voice signals, thus increasing the complexity and difficulty of use.
In summary, in the prior art, in the multi-terminal connection state of the teleconference, the parameter adjustment for suppressing the noise is relatively complex, and the technical problem of real-time communication requirement cannot be satisfied.
Disclosure of Invention
The application aims to solve the technical problems that in the prior art, parameter adjustment for suppressing noise is relatively complex in a multi-terminal connection state of a teleconference, and real-time communication requirements cannot be met by providing the intelligent noise reduction method and system for teleconference communication.
In view of the above problems, the present application provides a method and system for intelligent noise reduction in teleconference communication.
In a first aspect of the disclosure, a method for intelligent noise reduction in teleconferencing is provided, where the method is applied to an intelligent noise reduction system, the system includes a microphone array, a voice recognition module, a voice enhancement module, and a play module, and the method includes: before a target conference room executes a conference, acquiring an audio signal of the target conference room by using the microphone array, and acquiring a first environment signal; during the conference execution process of the target conference room, acquiring an audio signal of the target conference room by using the microphone array, and acquiring an audio preprocessing signal, wherein the audio preprocessing signal comprises a steady noise signal and a non-steady noise signal; the method comprises the steps of obtaining meeting personnel information of the target meeting room, wherein the meeting personnel information comprises voice information of the meeting personnel; the voice recognition module is used for carrying out voice recognition on the information of the participants, and recognizing voice signals of the participants; based on the voice signal, the voice enhancement module is used for enhancing the audio preprocessing signal to obtain an audio primary processing signal; after the target meeting room sends out a remote connection request, acquiring an audio signal of a remote connection end by using the microphone array, and acquiring a second environment signal; and performing multi-source noise elimination through the first environment signal, the second environment signal and the audio primary processing signal, obtaining an audio processing signal, and sending the audio processing signal to the playing module for playing.
In another aspect of the disclosure, a teleconferencing communication intelligent noise reduction system is provided, wherein the system includes a microphone array, a voice recognition module, a voice enhancement module, a play module, the system includes: the first environment signal acquisition module is used for acquiring an audio signal of a target conference room by using the microphone array before the conference is executed in the target conference room, so as to acquire a first environment signal; the audio signal acquisition module is used for acquiring an audio signal of the target conference room by using the microphone array in the conference execution process of the target conference room to acquire an audio preprocessing signal, wherein the audio preprocessing signal comprises a steady noise signal and a non-steady noise signal; the meeting personnel information acquisition module is used for acquiring meeting personnel information of the target meeting room, wherein the meeting personnel information comprises voice information of the meeting personnel; the voice recognition module is used for carrying out voice recognition on the information of the participants through the voice recognition module and recognizing voice signals of the participants; the enhancement processing module is used for carrying out enhancement processing on the audio preprocessing signal by using the voice enhancement module based on the voice signal to obtain an audio primary processing signal; the second environment signal acquisition module is used for acquiring an audio signal of a remote connection end by using the microphone array after the target conference room sends a remote connection request to acquire a second environment signal; and the multi-source noise elimination module is used for carrying out multi-source noise elimination through the first environment signal, the second environment signal and the audio primary processing signal, obtaining an audio processing signal and sending the audio processing signal to the playing module for playing.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
because the first environment signal and the audio preprocessing signal are acquired by utilizing the microphone array to acquire the audio signals before and during the meeting of the target meeting room, the signals comprise steady state noise signals and non-steady noise signals. Meanwhile, the voice information of the participants is acquired, and the voice signal is recognized through the voice recognition module. And based on the voice signal, performing enhancement processing on the audio pre-processing signal by using a voice enhancement module to obtain an audio primary processing signal. And after the target meeting room sends out the remote connection request, acquiring an audio signal of the remote connection end to acquire a second environment signal. The multi-source noise elimination is carried out by combining the first environment signal, the second environment signal and the audio primary processing signal, the audio processing signal is obtained and played, the interference of various noises inside and outside a conference room is effectively reduced, the voice quality is improved, multi-source voice with noise is processed in real time in a multi-terminal connection state of a remote conference, and conference participants can hear the voices of all parties at different connection terminals more clearly, so that the technical effects of the efficiency and the quality of the conference are improved.
The foregoing description is only an overview of the technical solutions of the present application, and may be implemented according to the content of the specification in order to make the technical means of the present application more clearly understood, and in order to make the above-mentioned and other objects, features and advantages of the present application more clearly understood, the following detailed description of the present application will be given.
Drawings
Fig. 1 is a schematic flow chart of a possible method for intelligent noise reduction in teleconference according to an embodiment of the present application;
fig. 2 is a schematic flow chart of a possible process of obtaining a neutralizing noise sound wave in the intelligent noise reduction method for teleconference communication according to the embodiment of the present application;
fig. 3 is a schematic diagram of a possible structure of a teleconference communication intelligent noise reduction system according to an embodiment of the present application.
Reference numerals illustrate: the system comprises a first environment signal acquisition module 100, an audio signal acquisition module 200, a participant information acquisition module 300, a voice recognition module 400, an enhancement processing module 500, a second environment signal acquisition module 600 and a multi-source noise elimination module 700.
Detailed Description
Exemplary embodiments of the present invention will now be described with reference to the accompanying drawings, in which various details of the embodiments of the present invention are included to facilitate understanding, and are to be considered merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the invention. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
Example 1
As shown in fig. 1, an embodiment of the present application provides a teleconference communication intelligent noise reduction method, where the method is applied to an intelligent noise reduction system, the system includes a microphone array, a voice recognition module, a voice enhancement module, and a play module, and the method includes:
step-1: before a target conference room executes a conference, acquiring an audio signal of the target conference room by using the microphone array, and acquiring a first environment signal;
step-2: during the conference execution process of the target conference room, acquiring an audio signal of the target conference room by using the microphone array, and acquiring an audio preprocessing signal, wherein the audio preprocessing signal comprises a steady noise signal and a non-steady noise signal;
step-3: the method comprises the steps of obtaining meeting personnel information of the target meeting room, wherein the meeting personnel information comprises voice information of the meeting personnel;
the system comprises a microphone array, a voice recognition module, a voice enhancement module and a play module, wherein the microphone array, the voice recognition module, the voice enhancement module and the play module are in communication connection, and the communication connection is simply through signal transmission interaction, and communication is formed among the microphone array, the voice recognition module, the voice enhancement module and the play module;
before the target conference room executes the conference, the microphone array is used for collecting the audio signals of the target conference room, including background noise such as indoor environment sound and air conditioning sound, so that the background noise condition of the target conference room can be known, and a reference is provided for subsequent noise processing.
During the conference execution process of the target conference room, the microphone array is used for collecting the audio signals again, and the audio preprocessing signals are obtained, wherein the audio preprocessing signals comprise steady-state noise signals (such as continuous background noise like air conditioning sound) and unsteady-state noise signals (such as footstep sound, applause and the like). For both types of noise signals, processing can be performed separately.
The method comprises the steps of acquiring the participant information of the target meeting room, including the voice information of the participant, so as to better know the speaking condition of the participant, so that the voice recognition and enhancement processing are more accurate, and data support is provided for subsequent analysis.
It should be noted that the noise reduction method for teleconference communication is mainly based on a signal processing technology, and the noise is suppressed by analyzing and processing an audio signal, and the specific method is as follows: the Kalman filtering method is a noise reduction method based on state estimation, and is suitable for filtering steady-state noise signals in the audio preprocessing signals by estimating the state of the voice signals and filtering estimation errors to obtain purer voice signals; the wiener filtering method is based on a minimum mean square error criterion, by estimating the power spectrum of the voice with noise and the noise, calculating an optimal filtering coefficient, and then applying the filtering coefficient to the voice with noise to obtain a purer voice signal, and is suitable for filtering the unsteady noise signal in the audio preprocessing signal.
Step-4: the voice recognition module is used for carrying out voice recognition on the information of the participants, and recognizing voice signals of the participants;
step-5: based on the voice signal, the voice enhancement module is used for enhancing the audio preprocessing signal to obtain an audio primary processing signal;
step-6: after the target meeting room sends out a remote connection request, acquiring an audio signal of a remote connection end by using the microphone array, and acquiring a second environment signal;
step-7: and performing multi-source noise elimination through the first environment signal, the second environment signal and the audio primary processing signal, obtaining an audio processing signal, and sending the audio processing signal to the playing module for playing.
The voice recognition module is used for carrying out voice recognition on the information of the participants, the recognition of the voice signals of the participants can be realized through a voice recognition technology, and the voice information of the participants is commonly recognized and extracted by using a voice recognition model based on deep learning; based on the voice signal, the voice enhancement module is used for enhancing the audio preprocessing signal to obtain an audio primary processing signal, and noise reduction, reverberation removal and the like can be carried out on the audio signal so as to improve the definition and audibility of the audio signal.
After the target meeting room sends a remote connection request, the microphone array is used for collecting audio signals of a remote connection end to obtain second environment signals, and the second environment signals can be obtained by analyzing the audio signals of the remote connection end, for example, information such as noise types, noise intensities and the like in the audio signals is analyzed; the audio processing signal is acquired by subjecting the first environmental signal, the second environmental signal, and the audio primary processing signal to a multi-source noise cancellation process, such as using a multi-source noise cancellation algorithm based on beam forming and deep learning to cancel the effects of multiple noise sources.
And sending the acquired audio processing signals to the playing module for playing. The processed audio signals are output to the participants through the playing module, so that the participants can better listen to conference contents. The audio signal is processed in the conference process, the influence of a plurality of noise sources is eliminated, and the definition and the audibility of the audio signal are improved, so that the smooth progress of the conference and the receiving and understanding of conference contents by participants are better ensured.
After the target meeting room sends out the remote connection request, the embodiment of the application further comprises the following steps:
performing acoustic environment analysis on the second environment signal, and positioning a second noise source position coordinate of the remote connection end;
based on the second noise source position coordinates, respectively acquiring the sound volume, and if the sound volume exceeds a noise volume threshold, marking the second noise source position coordinates corresponding to the sound volume exceeding the noise volume threshold by using a noise warning mark;
and identifying the noise type corresponding to the position coordinates of the second noise source with the noise warning mark, and judging whether a noise prohibiting instruction needs to be sent out or not according to the noise type.
After the target meeting room sends out the remote connection request, the method further comprises the following steps: and performing acoustic environment analysis on the second environment signal, wherein the acoustic environment analysis comprises the step of analyzing the acquired second environment signal by utilizing an acoustic analysis algorithm to locate the position coordinates of a second noise source of the remote connection end, and the position coordinates of the noise source generated by the remote connection end can be determined through the acoustic environment analysis.
Based on the second noise source position coordinates, respectively acquiring volume, and acquiring volume information of the position according to the second noise source position coordinates; and judging whether the sound volume exceeds a noise volume threshold value. If the volume exceeds the threshold (by default 70 dB), the second noise source location coordinates corresponding to the volume exceeding the noise volume threshold are marked with a noise alert flag, which may be an audible cue that the volume for a location exceeds the threshold.
And finally, identifying the noise type corresponding to the second noise source position coordinate with the noise warning mark, and judging whether a noise prohibiting instruction needs to be sent out or not according to the noise type. If the noise type is identified as a noise type that needs to be prohibited (e.g., if the noise type is speech or music), then a noise prohibit command is issued, which may be to a device at the remote connection to reduce the volume or stop generating noise, or to a device at the target conference room to filter or mask noise generated by the remote connection.
When the remote connection end generates noise, the remote connection end is positioned, the volume is detected and the type is identified, and a noise prohibiting instruction is sent out when necessary, so that noise interference generated by the remote connection end is effectively controlled, and the communication quality of a target conference room and the conference experience of participants are improved.
The embodiment of the application further comprises:
acquiring conference theme information of the target conference room for executing the conference, wherein the conference theme information comprises conference core content, conference time and conference place;
carrying out correlation analysis on the noise type and conference core content in the conference subject information to obtain noise correlation degree;
judging whether a noise prohibiting instruction needs to be sent out or not according to the noise correlation degree, if yes, taking the meeting time as supplementary information and storing the supplementary information into the noise prohibiting instruction.
Acquiring conference theme information of a target conference room for executing a conference, wherein the conference theme information comprises conference core content, conference time, conference place and the like, and can be acquired from conference organizers or participants or extracted by analyzing audio signals of the conference room; the noise correlation degree can be obtained by carrying out correlation analysis on the noise types and the conference core content in the conference subject information, the noise correlation degree is realized by counting and analyzing the occurrence frequency of various noise types when the conference core content appears, and particularly, the noise types in the target conference room can be classified and counted, such as human voice, air conditioning voice, environmental noise and the like; when the conference core content appears, recording or video of the noise types, and analyzing the recording or video by using a voice recognition technology to count the occurrence frequency of various noise types when the conference core content appears; by the method, the noise types which have larger interference to the conference can be determined more accurately, so that corresponding noise reduction measures are adopted, for example, noise reduction measures such as adding sound insulation materials in a region with larger noise, changing the internal structure of the conference room and the like can be adopted, or meeting participants can be reminded or warned to reduce noise of certain types and the like.
Judging whether a noise prohibiting instruction needs to be sent out or not through the noise correlation degree: if the noise correlation exceeds a certain threshold, the inertia of the noise and the conference core content is larger, and if a noise prohibiting instruction is sent, the description of the conference core content can be influenced, for example, the conference subject itself is to study the noise reduction of an engine, the noise of the engine is directly shielded, and the conference subject cannot be intuitively understood;
when the noise prohibiting instruction is issued, the conference time may be saved as supplementary information in the instruction, so that the device or system receiving the instruction may automatically perform the noise prohibiting operation, such as suspending the audio output of the remote connection terminal device or lowering the volume, etc., at a designated time. Judging whether interference noise is needed according to the correlation of the conference subject information and the noise type, and automatically executing a noise prohibiting instruction when necessary, so that smooth progress of the conference and understanding of conference contents by participants are better ensured.
The embodiment of the application further comprises:
performing acoustic environment analysis on the first environment signal, and positioning a first noise source position coordinate of the target conference room;
calculating noise source distance information through the meeting place and the first noise source position coordinates;
and if the noise source distance information is in the noise jurisdiction range, sending out a noise prohibiting instruction.
And carrying out acoustic environment analysis on the first environment signal, positioning the position coordinates of the first noise source of the target conference room, analyzing the audio signal of the conference room, and determining the position coordinates of the noise source by utilizing an acoustic analysis algorithm. Noise source distance information may be calculated from the conference site and the first noise source position coordinates, may be achieved by measuring the distance between the conference site and the noise source, or may be determined by analyzing the audio signal.
If the noise source distance information is within a noise jurisdiction (e.g., the subject division of the target meeting room is defined as the noise jurisdiction, which is not at maximum beyond the subject company of the target meeting room), a noise prohibit command may be issued and implemented by sending a command to the device at the remote connection, which may include the noise prohibit command and meeting location information. Noise source distance information is calculated according to meeting places and noise source position information, and a noise prohibiting instruction is automatically sent out when necessary, so that smooth meeting and meeting participants receiving meeting contents are better ensured.
As shown in fig. 2, if the noise source distance information is not in the noise jurisdiction, the embodiment of the present application further includes:
respectively measuring noise data from the inside to the outside of the target conference room in a time interval under the condition that the noise shielding measures of the target conference room are in a normal operation state, and measuring the equivalent sound level;
judging whether meeting the sound insulation upgrading standard of the conference room or not according to the equivalent sound level and the environmental noise basic standard;
and if the sound insulation upgrading standard of the conference room is not met, the noise is neutralized by adopting an active noise reduction technology, and the noise-neutralized sound wave is obtained.
When the noise shielding measures of the target conference room are in a normal running state, noise data measurement is carried out on the interior and the exterior of the target conference room in time intervals, the equivalent sound level is measured, the noise data measurement can be realized by arranging noise measurement equipment in the target conference room, and the noise measurement equipment can automatically or manually carry out the noise data measurement and record; by comparing the equivalent sound level with the basic standard of the environmental noise, whether meeting the sound insulation upgrading standard of the conference room (meeting the sound insulation design specification of civil construction of GB 50118-2010) can be judged, and the judgment can be carried out according to the specific environmental noise standard and the sound insulation upgrading standard of the conference room.
If the judging result is that the sound insulation upgrading standard of the conference room is not met, the noise can be neutralized by adopting an active noise reduction technology, the noise can be neutralized, the sound wave can be obtained by using an active noise reduction device or algorithm, and the active noise reduction device or algorithm can generate opposite sound waves according to the type and the characteristics of the noise, so that the noise can be neutralized.
When the noise source distance information is not in the noise jurisdiction range, the equivalent sound level is measured and compared with the environmental noise basic standard to judge whether the conference room sound insulation upgrading is needed, and an active noise reduction technology is adopted to neutralize noise when necessary, so that the smooth proceeding of the conference and the receiving of conference contents by participants are better ensured.
If the noise source distance information is in the noise jurisdiction, the embodiment of the application further includes:
in the noise jurisdiction, taking the target conference room as a center and the noise source distance information as a radius, carrying out acoustic environment layout modeling to obtain an acoustic model of the conference room;
dividing a distribution area of the first noise source position coordinates based on the conference room acoustic model to obtain a noise source concentrated distribution division area and a noise source independent distribution division area;
a centralized noise reduction measure is added in the noise source centralized distribution division area, and the centralized noise reduction measure accords with the environmental noise basic standard;
and in the independent distribution dividing area of the noise source, adding individual noise reduction measures, wherein the individual noise reduction measures accord with the basic standard of the environmental noise.
If the noise source distance information is within the noise jurisdiction, further comprising: taking a target conference room as a center, taking noise source distance information as a radius, carrying out acoustic environment layout modeling, and obtaining a conference room acoustic model, wherein the acoustic environment layout modeling can be realized by using acoustic simulation software or algorithm, and can comprise analysis and simulation of internal structures and materials of the conference room, the position and characteristics of a noise source, the path and effect of sound propagation and other factors.
Based on the conference room acoustic model, the distribution area of the first noise source position coordinate can be divided, so that a noise source concentrated distribution division area and a noise source independent distribution division area are obtained, and the division can be performed according to factors such as position information in the acoustic model, characteristics of the noise source, a sound propagation path and the like.
The noise sources are intensively distributed and divided into areas, and centralized noise reduction measures can be added, wherein the centralized noise reduction measures can comprise the use of sound insulation materials, the change of the internal structure of a conference room, the use of a muffler and the like, and the centralized noise reduction measures are in accordance with the basic standard of environmental noise so as to achieve the purpose of reducing noise.
In the area where the noise sources are independently distributed and divided, individual noise reduction measures can be added, wherein the individual noise reduction measures can comprise measures taken for each independent noise source, such as using personal protective equipment (such as equipment sound insulation pad and the like) or isolating or eliminating the independent noise source, and the individual noise reduction measures also accord with the basic standard of environmental noise so as to achieve the purpose of reducing noise.
When the noise source distance information is in the noise jurisdiction, modeling and dividing the acoustic environment layout are carried out, and targeted noise reduction measures are adopted, so that smooth progress of a conference and conference contents of participants are better ensured.
The centralized noise reduction measure accords with the environmental noise basic standard, and the embodiment of the application further comprises:
determining a noise reduction device model and the number of noise reduction devices according to the noise volume level, the first noise source position coordinates, the noise source distance information, the environmental noise basic standard and the centralized noise reduction measure;
acquiring a noise reduction device operation parameter value interval based on the noise reduction device model number and the noise reduction device number;
marking a noise propagation critical path based on the conference room acoustic model, wherein the noise propagation critical path comprises a multi-section regional clustering noise reduction equipment operation parameter value interval;
noise reduction equipment is installed within the noise jurisdiction based on the noise propagation critical path.
The concentrated noise reduction measures should meet the ambient noise base standard. In determining the model number and the number of the noise reduction devices, consideration and calculation are required according to the noise volume level, the first noise source position coordinates, the noise source distance information, the environmental noise basic standard and the centralized noise reduction measures, for example, the model number and the number of the noise reduction devices to be used can be determined according to the noise volume level and the first noise source position coordinates; based on the model number and the number of the noise reduction devices, a value interval of the operation parameters of the noise reduction devices can be obtained, and the parameter range can be determined by inquiring the device use specification of the noise reduction devices.
Based on the conference room acoustic model, a noise propagation critical path can be marked, the noise propagation critical path can be performed by analyzing the sound propagation path and effect in the acoustic model and considering the operation parameter value range of the noise reduction device, the noise propagation critical path can comprise a noise reduction device operation parameter value interval of multi-section area clustering, and the noise reduction device operation parameter value interval can be divided and adjusted according to actual conditions.
Based on the noise propagation critical path, the noise reduction device may be installed in the noise jurisdiction, the noise reduction device may be installed by following the position information on the critical path, for example, selecting an appropriate position on the critical path to install the noise reduction device or adjusting the parameters of the noise reduction device according to the indication of the critical path, etc. The number of the noise reduction devices and the model number of the noise reduction devices are determined by taking various factors into consideration, and the key paths and actual application conditions are taken into consideration when the noise reduction devices are installed.
In summary, the method and system for intelligent noise reduction of teleconference communication provided by the embodiment of the application have the following technical effects:
1. the intelligent noise reduction method for teleconference communication can learn noise characteristics and voice characteristics from a large amount of data without presuming signal characteristics, so that non-stationary noise and complex noise can be better processed, and the noise reduction effect is improved.
2. The parameters are adjusted in a self-adaptive mode, manual setting and adjustment are not needed, and the complexity and difficulty of use are reduced.
3. By adopting the efficient and simple lightweight algorithm, real-time processing can be realized under lower computing resources and power consumption, and the requirement of real-time communication is met.
4. Because the acoustic environment layout modeling is carried out in the noise jurisdiction with the target meeting room as the center and the noise source distance information as the radius, the meeting room acoustic model is obtained; dividing a distribution area of the first noise source position coordinates based on the conference room acoustic model to obtain a noise source concentrated distribution division area and a noise source independent distribution division area; the noise sources are intensively distributed and divided into areas, and centralized noise reduction measures are additionally added, wherein the centralized noise reduction measures accord with the basic standard of environmental noise; in the independent distribution dividing area of the noise source, an individual noise reduction measure is additionally arranged, and the individual noise reduction measure accords with the basic standard of environmental noise. When the noise source distance information is in the noise jurisdiction, modeling and dividing the acoustic environment layout, and taking targeted noise reduction measures, so that smooth meeting and meeting participants receiving the meeting content are better ensured.
Example two
Based on the same inventive concept as the teleconference communication intelligent noise reduction method in the foregoing embodiment, as shown in fig. 3, an embodiment of the present application provides a teleconference communication intelligent noise reduction system, where the system includes a microphone array, a voice recognition module, a voice enhancement module, and a play module, and the system includes:
a first environmental signal acquisition module 100, configured to acquire, before a conference is performed in a target conference room, an audio signal of the target conference room by using the microphone array, and acquire a first environmental signal;
the audio signal acquisition module 200 is configured to acquire an audio signal of the target conference room by using the microphone array in a conference execution process of the target conference room, and acquire an audio preprocessing signal, where the audio preprocessing signal includes a stationary noise signal and a non-stationary noise signal;
the participant information acquisition module 300 is configured to acquire participant information of the target conference room, where the participant information includes participant voice information;
the voice recognition module 400 is configured to perform voice recognition on the participant information through the voice recognition module, and recognize a voice signal of the participant;
the enhancement processing module 500 is configured to perform enhancement processing on the audio pre-processing signal by using the speech enhancement module based on the speech signal, so as to obtain an audio primary processing signal;
the second environmental signal acquisition module 600 is configured to acquire, after the target conference room sends a remote connection request, an audio signal of a remote connection end by using the microphone array, and acquire a second environmental signal;
the multi-source noise cancellation module 700 is configured to perform multi-source noise cancellation by using the first environmental signal, the second environmental signal, and the audio primary processing signal, obtain an audio processing signal, and send the audio processing signal to the playing module for playing.
Further, after the target conference room issues a remote connection request, the system further includes the following steps:
performing acoustic environment analysis on the second environment signal, and positioning a second noise source position coordinate of the remote connection end;
based on the second noise source position coordinates, respectively acquiring the sound volume, and if the sound volume exceeds a noise volume threshold, marking the second noise source position coordinates corresponding to the sound volume exceeding the noise volume threshold by using a noise warning mark;
and identifying the noise type corresponding to the position coordinates of the second noise source with the noise warning mark, and judging whether a noise prohibiting instruction needs to be sent out or not according to the noise type.
Further, the system further comprises the following steps:
acquiring conference theme information of the target conference room for executing the conference, wherein the conference theme information comprises conference core content, conference time and conference place;
carrying out correlation analysis on the noise type and conference core content in the conference subject information to obtain noise correlation degree;
judging whether a noise prohibiting instruction needs to be sent out or not according to the noise correlation degree, if yes, taking the meeting time as supplementary information and storing the supplementary information into the noise prohibiting instruction.
Further, the system further comprises the following steps:
performing acoustic environment analysis on the first environment signal, and positioning a first noise source position coordinate of the target conference room;
calculating noise source distance information through the meeting place and the first noise source position coordinates;
and if the noise source distance information is in the noise jurisdiction range, sending out a noise prohibiting instruction.
Further, the system further comprises the following steps:
if the noise source distance information is not in the noise jurisdiction range, respectively measuring noise data from the inside to the outside of the target conference room in a time interval under the normal running state of the noise shielding measure of the target conference room, and measuring the equivalent sound level;
judging whether meeting the sound insulation upgrading standard of the conference room or not according to the equivalent sound level and the environmental noise basic standard;
and if the sound insulation upgrading standard of the conference room is not met, the noise is neutralized by adopting an active noise reduction technology, and the noise-neutralized sound wave is obtained.
Further, if the noise source distance information is within the noise jurisdiction, the system further comprises the steps of:
in the noise jurisdiction, taking the target conference room as a center and the noise source distance information as a radius, carrying out acoustic environment layout modeling to obtain an acoustic model of the conference room;
dividing a distribution area of the first noise source position coordinates based on the conference room acoustic model to obtain a noise source concentrated distribution division area and a noise source independent distribution division area;
a centralized noise reduction measure is added in the noise source centralized distribution division area, and the centralized noise reduction measure accords with the environmental noise basic standard;
and in the independent distribution dividing area of the noise source, adding individual noise reduction measures, wherein the individual noise reduction measures accord with the basic standard of the environmental noise.
Further, the centralized noise reduction measure meets the environmental noise basic standard, and the system further comprises the following steps:
determining a noise reduction device model and the number of noise reduction devices according to the noise volume level, the first noise source position coordinates, the noise source distance information, the environmental noise basic standard and the centralized noise reduction measure;
acquiring a noise reduction device operation parameter value interval based on the noise reduction device model number and the noise reduction device number;
marking a noise propagation critical path based on the conference room acoustic model, wherein the noise propagation critical path comprises a multi-section regional clustering noise reduction equipment operation parameter value interval;
noise reduction equipment is installed within the noise jurisdiction based on the noise propagation critical path.
Any of the steps of the methods described above may be stored as computer instructions or programs in a non-limiting computer memory and may be called by a non-limiting computer processor to identify any of the methods to implement embodiments of the present application, without unnecessary limitations.
Further, the first or second element may not only represent a sequential relationship, but may also represent a particular concept, and/or may be selected individually or in whole among a plurality of elements. It will be apparent to those skilled in the art that various modifications and variations can be made in the present application without departing from the scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the present application and the equivalents thereof, the present application is intended to cover such modifications and variations.
Claims (8)
1. The intelligent noise reduction method for teleconference communication is characterized by being applied to an intelligent noise reduction system, wherein the system comprises a microphone array, a voice recognition module, a voice enhancement module and a play module, and the method comprises the following steps:
before a target conference room executes a conference, acquiring an audio signal of the target conference room by using the microphone array, and acquiring a first environment signal;
during the conference execution process of the target conference room, acquiring an audio signal of the target conference room by using the microphone array, and acquiring an audio preprocessing signal, wherein the audio preprocessing signal comprises a steady noise signal and a non-steady noise signal;
the method comprises the steps of obtaining meeting personnel information of the target meeting room, wherein the meeting personnel information comprises voice information of the meeting personnel;
the voice recognition module is used for carrying out voice recognition on the information of the participants, and recognizing voice signals of the participants;
based on the voice signal, the voice enhancement module is used for enhancing the audio preprocessing signal to obtain an audio primary processing signal;
after the target meeting room sends out a remote connection request, acquiring an audio signal of a remote connection end by using the microphone array, and acquiring a second environment signal;
and performing multi-source noise elimination through the first environment signal, the second environment signal and the audio primary processing signal, obtaining an audio processing signal, and sending the audio processing signal to the playing module for playing.
2. The method of claim 1, wherein after the target conference room issues a remote connection request, the method further comprises:
performing acoustic environment analysis on the second environment signal, and positioning a second noise source position coordinate of the remote connection end;
based on the second noise source position coordinates, respectively acquiring the sound volume, and if the sound volume exceeds a noise volume threshold, marking the second noise source position coordinates corresponding to the sound volume exceeding the noise volume threshold by using a noise warning mark;
and identifying the noise type corresponding to the position coordinates of the second noise source with the noise warning mark, and judging whether a noise prohibiting instruction needs to be sent out or not according to the noise type.
3. The method of claim 2, wherein the method further comprises:
acquiring conference theme information of the target conference room for executing the conference, wherein the conference theme information comprises conference core content, conference time and conference place;
carrying out correlation analysis on the noise type and conference core content in the conference subject information to obtain noise correlation degree;
judging whether a noise prohibiting instruction needs to be sent out or not according to the noise correlation degree, if yes, taking the meeting time as supplementary information and storing the supplementary information into the noise prohibiting instruction.
4. A method as recited in claim 3, further comprising:
performing acoustic environment analysis on the first environment signal, and positioning a first noise source position coordinate of the target conference room;
calculating noise source distance information through the meeting place and the first noise source position coordinates;
and if the noise source distance information is in the noise jurisdiction range, sending out a noise prohibiting instruction.
5. The method of claim 4, wherein if the noise source distance information is not in the noise jurisdiction,
respectively measuring noise data from the inside to the outside of the target conference room in a time interval under the condition that the noise shielding measures of the target conference room are in a normal operation state, and measuring the equivalent sound level;
judging whether meeting the sound insulation upgrading standard of the conference room or not according to the equivalent sound level and the environmental noise basic standard;
and if the sound insulation upgrading standard of the conference room is not met, the noise is neutralized by adopting an active noise reduction technology, and the noise-neutralized sound wave is obtained.
6. The method of claim 5, wherein if the noise source distance information is within the noise jurisdiction, the method further comprises:
in the noise jurisdiction, taking the target conference room as a center and the noise source distance information as a radius, carrying out acoustic environment layout modeling to obtain an acoustic model of the conference room;
dividing a distribution area of the first noise source position coordinates based on the conference room acoustic model to obtain a noise source concentrated distribution division area and a noise source independent distribution division area;
a centralized noise reduction measure is added in the noise source centralized distribution division area, and the centralized noise reduction measure accords with the environmental noise basic standard;
and in the independent distribution dividing area of the noise source, adding individual noise reduction measures, wherein the individual noise reduction measures accord with the basic standard of the environmental noise.
7. The method of claim 6, wherein the centralized noise reduction measure meets the ambient noise floor criteria, the method further comprising:
determining a noise reduction device model and the number of noise reduction devices according to the noise volume level, the first noise source position coordinates, the noise source distance information, the environmental noise basic standard and the centralized noise reduction measure;
acquiring a noise reduction device operation parameter value interval based on the noise reduction device model number and the noise reduction device number;
marking a noise propagation critical path based on the conference room acoustic model, wherein the noise propagation critical path comprises a multi-section regional clustering noise reduction equipment operation parameter value interval;
noise reduction equipment is installed within the noise jurisdiction based on the noise propagation critical path.
8. A teleconferencing intelligent noise reduction system for implementing the teleconferencing intelligent noise reduction method of any of claims 1-7, the system comprising a microphone array, a speech recognition module, a speech enhancement module, a playback module, comprising:
the first environment signal acquisition module is used for acquiring an audio signal of a target conference room by using the microphone array before the conference is executed in the target conference room, so as to acquire a first environment signal;
the audio signal acquisition module is used for acquiring an audio signal of the target conference room by using the microphone array in the conference execution process of the target conference room to acquire an audio preprocessing signal, wherein the audio preprocessing signal comprises a steady noise signal and a non-steady noise signal;
the meeting personnel information acquisition module is used for acquiring meeting personnel information of the target meeting room, wherein the meeting personnel information comprises voice information of the meeting personnel;
the voice recognition module is used for carrying out voice recognition on the information of the participants through the voice recognition module and recognizing voice signals of the participants;
the enhancement processing module is used for carrying out enhancement processing on the audio preprocessing signal by using the voice enhancement module based on the voice signal to obtain an audio primary processing signal;
the second environment signal acquisition module is used for acquiring an audio signal of a remote connection end by using the microphone array after the target conference room sends a remote connection request to acquire a second environment signal;
and the multi-source noise elimination module is used for carrying out multi-source noise elimination through the first environment signal, the second environment signal and the audio primary processing signal, obtaining an audio processing signal and sending the audio processing signal to the playing module for playing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410185685.5A CN117746880A (en) | 2024-02-20 | 2024-02-20 | Intelligent noise reduction method and system for teleconference communication |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410185685.5A CN117746880A (en) | 2024-02-20 | 2024-02-20 | Intelligent noise reduction method and system for teleconference communication |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117746880A true CN117746880A (en) | 2024-03-22 |
Family
ID=90259479
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410185685.5A Pending CN117746880A (en) | 2024-02-20 | 2024-02-20 | Intelligent noise reduction method and system for teleconference communication |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117746880A (en) |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050094795A1 (en) * | 2003-10-29 | 2005-05-05 | Broadcom Corporation | High quality audio conferencing with adaptive beamforming |
| CN102890936A (en) * | 2011-07-19 | 2013-01-23 | 联想(北京)有限公司 | Audio processing method and terminal device and system |
| CN104580776A (en) * | 2015-01-16 | 2015-04-29 | 四川联友电讯技术有限公司 | Telephone conference system and method capable of intelligently shielding strong noise participant based on noise detection |
| CN105933556A (en) * | 2016-04-25 | 2016-09-07 | 四川联友电讯技术有限公司 | Teleconference noise shielding system based on conference participant determination and teleconference noise shielding method thereof |
| CN108520754A (en) * | 2018-04-09 | 2018-09-11 | 广东思派康电子科技有限公司 | A kind of noise reduction meeting machine |
| CN109920434A (en) * | 2019-03-11 | 2019-06-21 | 南京邮电大学 | A kind of noise classification minimizing technology based on conference scenario |
| CN110300001A (en) * | 2019-05-21 | 2019-10-01 | 深圳壹账通智能科技有限公司 | Conference audio control method, system, equipment and computer readable storage medium |
| US20200110572A1 (en) * | 2018-10-08 | 2020-04-09 | Nuance Communications, Inc. | System and method for managing a mute button setting for a conference call |
| CN111554314A (en) * | 2020-05-15 | 2020-08-18 | 腾讯科技（深圳）有限公司 | Noise detection method, device, terminal and storage medium |
| CN114171041A (en) * | 2021-11-30 | 2022-03-11 | 深港产学研基地（北京大学香港科技大学深圳研修院） | Voice noise reduction method, device and equipment based on environment detection and storage medium |
| CN114792524A (en) * | 2022-06-24 | 2022-07-26 | 腾讯科技（深圳）有限公司 | Audio data processing method, apparatus, program product, computer device and medium |
-
2024
- 2024-02-20 CN CN202410185685.5A patent/CN117746880A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050094795A1 (en) * | 2003-10-29 | 2005-05-05 | Broadcom Corporation | High quality audio conferencing with adaptive beamforming |
| CN102890936A (en) * | 2011-07-19 | 2013-01-23 | 联想(北京)有限公司 | Audio processing method and terminal device and system |
| CN104580776A (en) * | 2015-01-16 | 2015-04-29 | 四川联友电讯技术有限公司 | Telephone conference system and method capable of intelligently shielding strong noise participant based on noise detection |
| CN105933556A (en) * | 2016-04-25 | 2016-09-07 | 四川联友电讯技术有限公司 | Teleconference noise shielding system based on conference participant determination and teleconference noise shielding method thereof |
| CN108520754A (en) * | 2018-04-09 | 2018-09-11 | 广东思派康电子科技有限公司 | A kind of noise reduction meeting machine |
| US20200110572A1 (en) * | 2018-10-08 | 2020-04-09 | Nuance Communications, Inc. | System and method for managing a mute button setting for a conference call |
| CN109920434A (en) * | 2019-03-11 | 2019-06-21 | 南京邮电大学 | A kind of noise classification minimizing technology based on conference scenario |
| CN110300001A (en) * | 2019-05-21 | 2019-10-01 | 深圳壹账通智能科技有限公司 | Conference audio control method, system, equipment and computer readable storage medium |
| CN111554314A (en) * | 2020-05-15 | 2020-08-18 | 腾讯科技（深圳）有限公司 | Noise detection method, device, terminal and storage medium |
| CN114171041A (en) * | 2021-11-30 | 2022-03-11 | 深港产学研基地（北京大学香港科技大学深圳研修院） | Voice noise reduction method, device and equipment based on environment detection and storage medium |
| CN114792524A (en) * | 2022-06-24 | 2022-07-26 | 腾讯科技（深圳）有限公司 | Audio data processing method, apparatus, program product, computer device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11569789B2 (en) | Compensation for ambient sound signals to facilitate adjustment of an audio volume | |
| CN110503969B (en) | Audio data processing method and device and storage medium | |
| Benesty et al. | Speech enhancement | |
| CN107910011B (en) | Voice noise reduction method and device, server and storage medium | |
| CN102388416B (en) | Signal processing apparatus and signal processing method | |
| CN103152546B (en) | Based on pattern recognition and the video conference echo suppressing method postponing feedfoward control | |
| CN103413547B (en) | A kind of method that room reverberation is eliminated | |
| CN102884575A (en) | Voice activity detection | |
| CN104246877A (en) | Systems and methods for audio signal processing | |
| CN103440869A (en) | Audio-reverberation inhibiting device and inhibiting method thereof | |
| CN104658543A (en) | Method for eliminating indoor reverberation | |
| CN110830901B (en) | Multichannel sound amplifying system and method for adjusting volume of loudspeaker | |
| CN110970010A (en) | Noise elimination method, device, storage medium and equipment | |
| CN113241085B (en) | Echo cancellation method, device, equipment and readable storage medium | |
| CN110956976A (en) | Echo cancellation method, device, equipment and readable storage medium | |
| CN110475181B (en) | Equipment configuration method, device, equipment and storage medium | |
| CN113949955A (en) | Noise reduction processing method and device, electronic equipment, earphone and storage medium | |
| CN112513977A (en) | Signal processing device and method, and program | |
| Gamper et al. | Predicting word error rate for reverberant speech | |
| CN112333602B (en) | Signal processing method, signal processing apparatus, computer-readable storage medium, and indoor playback system | |
| CN117746880A (en) | Intelligent noise reduction method and system for teleconference communication | |
| CN105261363A (en) | Voice recognition method, device and terminal | |
| CN111412587A (en) | Voice processing method and device of air conditioner, air conditioner and storage medium | |
| US11890168B2 (en) | Hearing protection and situational awareness system | |
| CN116567513A (en) | Stereo earphone testing method and system based on filter |
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 |