TW201506913A - Microphone system and sound processing method thereof - Google Patents

Abstract

Disclosed herein is microphone system that includes a microphone, an acoustic echo cancellation unit, a noise reduction unit, a gain control unit and an equalization unit. The acoustic echo cancelation unit is configured to cancel the sound signal played by the loudspeaker from a microphone signal to output a sound wave signal; the noise reduction unit is configured to filter out the background noise from the sound wave signal to output a denoised sound wave signal; the gain control is configured to compare a similarity between energy levels of the denoised sound wave signal and a reference model, and adjust the volume of the denoised sound wave signal based on the comparison result and output the volume-adjusted sound wave signal; the equalization unit is configured to compare a similarity between a signal-to-noise frequency energy of the volume-adjusted sound wave signal and a signal-to-noise frequency energy of the reference model, and execute an equalization process on the volume-adjusted sound wave signal based on the comparison result.

Description

Microphone system and sound processing method thereof

The present invention relates to a radio system, and more particularly to a microphone system and a sound processing method thereof.

With the advancement of technology, today's oral communication is paying more and more attention to the presentation of visual effects. Therefore, the presentations are often used in conjunction with lectures to provide a richer audience.

In addition to the basic infrared laser function, the projectors currently used in the market can be used by the speaker to indicate the key points of the speech, and also combine the function of page-changing pages, so that the speaker does not need to project in the middle of the lecture to the notebook or other storage device. The switching of the film increases the fluency of the speech process. However, due to the wide range of applications, ranging from small to small conference rooms and large lecture halls that can accommodate hundreds of people, the speaker has to use one hand to hold the projection pen for laser and slide switching. The hand also needs to use a microphone to expand the volume of the speech, which limits the scope of the speaker's actions and body expression. The body movements that the speaker needs to express may cause the distance between the microphone and the speaker to change constantly, causing the sound played by the speaker to flicker. Suddenly, the listener feels uncomfortable

It can be seen that the above-mentioned prior art obviously has inconveniences and defects, and needs to be further improved. In order to solve the above problems, the relevant fields have not exhausted their efforts to seek solutions, but the methods that have not been applied for a long time have been developed. Therefore, how to provide a more convenient speaking tool for users is one of the current important research and development topics, and it has become an urgent need for improvement in related fields.

One aspect of the present invention provides a microphone system and a sound processing method thereof, which can perform a series of processing on a microphone signal to achieve the effects of eliminating echo, reducing noise, adjusting volume and dynamic volume equivalent, and providing a speaker. Good experience and a comfortable feeling for the listener.

The microphone system provided by the present invention comprises a microphone, an echo cancellation unit, a noise attenuation unit, a volume control unit and a sound effect equalization unit. The echo cancellation unit removes the sound signal broadcasted by the speaker from the microphone signal received by the microphone to output the sound wave signal; the noise attenuation unit filters out the background noise of the microphone from the sound wave signal to output the noise signal of the noise reduction; volume control The unit compares the energy similarity between the denoised sound wave signal and the reference model, and accordingly adjusts the volume of the denoised sound wave signal and outputs the volume-adjusted sound wave signal; the sound effect equalizing unit compares the volume The degree of similarity between the signal-to-noise frequency energy ratio of the adjusted sound wave signal and the noise-to-noise frequency energy ratio of the reference model is performed, and the sound effect equalization process is performed on the volume-adjusted sound wave signal, and a processed sound wave signal is outputted. Give Yang Sounder.

In an embodiment, the echo cancellation unit determines a correlation between the microphone signal and the reference signal. When the correlation is lower than the preset value, the echo cancellation unit cancels the broadcast by the speaker according to the preset offset coefficient. Sound signal.

In an embodiment, when the correlation is higher than the preset value, the echo cancellation unit updates the offset coefficient based on the correlation, and eliminates the sound signal broadcast by the speaker according to the updated offset coefficient.

In an embodiment, the sound effect equalization unit calculates the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal, and determines whether the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to a preset equalization condition. When the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to a preset equalization condition, the sound effect equalization unit compares the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal with the reference noise-frequency energy of the reference model. The degree of similarity between the two, and the sound effect equalization processing is performed on the volume-adjusted sound wave signal according to the degree of similarity.

In an embodiment, the sound equalization unit records the processed sound wave signal as a reference signal before sending the processed sound wave signal to the speaker for playing.

In one embodiment, the reference model is based on acoustic characteristics of white noise collected by the microphone.

In an embodiment, the volume control unit determines whether the noise signal of the noise reduction conforms to a preset volume condition. When the noise signal of the noise reduction meets the preset volume condition, the volume control unit compares the sound signal with the noise reduction. Reference The energy similarity between the models is tested, and then the noise-reduced acoustic signals are volume-adjusted according to the energy similarity and the volume-adjusted acoustic signals are output.

In one embodiment, the noise attenuation unit subtracts the estimated background noise from the acoustic signal after estimating the background noise.

In an embodiment, the microphone system further includes a body and a laser pointer, wherein the body is configured to receive the microphone, and the laser indicator is disposed in the body.

In an embodiment, the microphone system further includes a remote controller disposed in the body for controlling the slide film played by the computer device.

The sound processing method is applicable to the microphone system, the method.

In another aspect, the sound processing method provided by the present invention comprises the steps of: (a) eliminating an audio signal broadcast by a speaker from a microphone signal received by a microphone to output a sound wave signal; (b) from the sound wave Filtering the background noise of the microphone to output a denoised sound wave signal; (c) comparing the energy similarity between the denoised sound wave signal and a reference model, according to which The noise sound signal is adjusted in volume and outputs a volume-adjusted sound wave signal; and (d) a ratio between the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal and the noise-to-noise frequency energy ratio of the reference model The degree of similarity is such that an audio effect equalization process is performed on the volume-adjusted sound wave signal and a processed sound wave signal is output to the speaker.

In an embodiment, the step (a) includes: determining a correlation between the microphone signal and a reference signal; and when the correlation is low When a preset value is used, the sound signal broadcast by the speaker is eliminated according to one of the preset offset coefficients.

In an embodiment, the step (a) further includes: when the correlation is higher than the preset value, updating the offset coefficient based on the correlation; and canceling the speaker according to the updated offset coefficient The sound signal that is being broadcast.

In an embodiment, the step (d) includes: calculating a noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal; determining whether the sound-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to a preset, etc. a condition; when the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to the preset equalization condition, the signal-to-noise frequency energy ratio of the volume-adjusted sound wave signal is compared with the noise model of the reference model a degree of similarity between the frequency energy ratios; and performing the sound equalization process on the volume-adjusted sound wave signal according to the degree of similarity.

In an embodiment, the step (d) further includes: recording the processed sound wave signal as the reference signal before sending the processed sound wave signal to the speaker for playing.

In an embodiment, the reference model is based on acoustic characteristics of white noise collected by the microphone.

In an embodiment, the step (c) includes: determining whether the denoised sound wave signal meets a preset volume condition; and when the denoised sound wave signal meets the preset volume condition, comparing The similarity of the energy between the noise-reduced acoustic wave signal and the reference model; and volume-adjusting the denoised sound wave signal according to the energy magnitude similarity and outputting the The sound signal after the volume adjustment.

In an embodiment, the step (b) includes: estimating the background noise; and deducting the estimated background noise from the sound wave signal.

In summary, the technical solution of the present invention has obvious advantages and beneficial effects compared with the prior art. With the above technical solution, considerable technological progress can be achieved, and the industrial use value is widely used, which has at least the following advantages: (1) combining the projection pen and the microphone into a single product, so that the speaker performs the slide film In the speech, you only need to hold the pen microphone with one hand to complete various functions such as laser, slide switch and sound reinforcement, so that the speaker's limb expression is not restricted by the equipment used; (2) Long-distance radio microphone The radio range is wide, eliminating the discomfort of the speaker due to the long-term use of the microphone; (3) The built-in sound processing method of the microphone system can instantly optimize the sound received by the microphone to provide the best auditory experience for the listener.

The above description will be described in detail in the following embodiments, and further explanation of the technical solutions of the present invention will be provided.

The above and other objects, features, advantages and embodiments of the present invention will become more apparent and understood.

100‧‧‧Microphone system

110‧‧‧ microphone

120‧‧‧Echo cancellation unit

130‧‧‧Noise attenuation unit

140‧‧‧Volume control unit

150‧‧‧Sound equalization unit

190‧‧‧Speaker

210‧‧‧ body

220‧‧‧Laser indicator

230‧‧‧Remote control

250‧‧‧Computer equipment

252‧‧‧Slide

290‧‧‧ Speaker

300‧‧‧Sound processing method

310‧‧‧Acoustic echo cancellation steps

320‧‧‧Noise attenuation step

330‧‧‧Automatic volume adjustment steps

340‧‧‧Dynamic sound equalization steps

350‧‧‧Reference signal recording steps

360‧‧‧Sounding steps

The above and other objects, features, advantages and embodiments of the present invention will become more <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; </ RTI> <RTIgt; Figure is a partial schematic view of a microphone system in accordance with an embodiment of the present invention; 3 is a flow chart of a sound processing method of a microphone system in accordance with an embodiment of the present invention.

In order to make the description of the present invention more complete and complete, reference is made to the accompanying drawings and the accompanying drawings. On the other hand, well-known elements and steps are not described in the embodiments to avoid unnecessarily limiting the invention.

In the scope of the embodiments and claims, the description of "coupled with" may refer to a component being indirectly connected to another component through other components, or a component may be directly connected to Another component.

1 is a block diagram of a microphone system 100 in accordance with an embodiment of the present invention. As shown in FIG. 1, the microphone system 100 includes a microphone 110, an echo cancellation unit 120, a noise attenuation unit 130, a volume control unit 140, and a sound effect equalization unit 150. The microphone 110 is coupled to the echo cancellation unit 120. The echo cancellation unit 120 is coupled to the noise attenuation unit 130. The noise attenuation unit 130 is coupled to the volume control unit 140. The volume control unit 140 is coupled to the audio equalization unit 150. The unit 150 is coupled to the speaker 190.

The echo cancellation unit 120 removes the sound signal broadcast by the speaker 190 from the microphone signal received by the microphone 110 to output the sound wave signal. Then, the noise attenuation unit 130 filters out the background noise of the microphone 110 from the sound wave signal to output the output. Noise sound signal; then, volume control The element 140 compares the energy similarity between the denoised sound wave signal and the reference model, and performs volume adjustment on the denoised sound wave signal and outputs the volume adjusted sound wave signal; then, the sound effect equalization unit 150 Comparing the similarity between the signal-to-noise frequency energy ratio of the volume-adjusted sound wave signal and the noise-to-noise frequency energy ratio of the reference model, performing sound equalization processing on the volume-adjusted sound wave signal, and outputting a processing The subsequent sound wave signal is given to the speaker 190, so that the speaker 190 plays the processed sound, thereby providing a better experience for the speaker and a comfortable feeling for the listener.

In practice, in order to achieve the effect of long-distance radio reception of the microphone 110, a monomer with higher sensitivity must be used in the selection of the microphone 110 to increase the range of the radio, and the sound emitted by the speaker 190 is more easily caused by the sensitivity of the microphone 110. The microphone 110 receives the sound signal, and the sound wave signal is received by the microphone 110 and then amplified and broadcasted through the speaker 190. This volume is continuously amplified between the microphone and the speaker to form a Howling phenomenon, which is called acoustic feedback, in order to avoid this phenomenon. The acoustic echo cancellation calculation is performed after the microphone 110 receives the acoustic signal, that is, the signal broadcasted by the microphone 110 and the speaker 190 is compared, and only the voice signal received by the "present" microphone 110 is played, and the "past" speaker 190 is played. Sound wave signal. Therefore, in an embodiment, the echo cancellation unit 120 determines the correlation between the microphone signal and the reference signal. When the correlation is lower than the preset value, the echo cancellation unit 120 cancels the speaker according to the preset offset coefficient. 190 sound signals broadcast. On the other hand, when the above correlation is higher than the preset value, the echo cancellation unit 120 updates the offset coefficient based on the correlation, and eliminates the sound played by the speaker 190 according to the updated offset coefficient. Signal.

Regarding the source of the reference signal, in an embodiment, the sound effect equalization unit 150 records the processed sound wave signal as the next echo calculation unit 120 before sending the processed sound wave signal to the speaker 190 for playing. And the "reference signal" used in the judgment.

On the other hand, the use of the highly sensitive microphone 110, in addition to the increased range of speech that can be received, also receives more background noise, so in one embodiment, the noise attenuation unit 130 is used to estimate the background. After the noise, the acoustic signal outputted by the echo cancellation unit 120 is subtracted from the estimated background noise to eliminate the background noise.

Moreover, the usage behavior set by the present invention does not limit the distance between the microphone 110 and the speaker, and the speaker's body motion required to express the expression may cause the distance between the microphone 110 and the speaker to constantly change, causing the speaker 190 to The sound played out is too big and small. In order to avoid such uncomfortable listener's feelings, it is necessary to make an automatic volume adjustment of the sound wave signal received by the microphone 110. Therefore, in an embodiment, the volume control unit 140 determines whether the denoised sound wave signal output by the noise attenuation unit 130 meets a preset volume condition, and when the noise reduced sound wave signal meets the preset volume condition, the volume control The unit 140 compares the energy similarity between the denoised sound wave signal and the reference model, and further adjusts the volume of the denoised sound wave signal according to the energy similarity degree and outputs the volume adjusted sound wave signal.

Regarding the source of the above reference model, in one embodiment, the reference model is based on the acoustic characteristics of the white noise collected by the microphone 110. or, In other embodiments, the reference model is based on various characteristics such as pink noise, gray noise collected by the microphone 110, and those skilled in the art should design the elastic design at that time.

Finally, since the sound signal is transmitted in the air, the energy is attenuated, and the energy attenuation of each frequency is different, which causes the distance between the sound source and the microphone 110 to be different, so that the listener has different hearing feelings. To solve this problem, it is necessary to join Sound signal processing for dynamic sound equalization. Therefore, in an embodiment, the sound effect equalization unit 150 calculates the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal, and determines whether the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to a preset equalization. The sound effect equalization unit 150 compares the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal with the noise of the reference model when the sound-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to a preset equalization condition. The degree of similarity between the frequency and energy ratios is performed on the volume-adjusted sound wave signal according to the degree of similarity, so that the listener has a better hearing experience.

2 is a partial schematic view of a microphone system in accordance with an embodiment of the present invention. As shown in FIG. 2 , the microphone system includes a body 210 and a laser pointer 220 for receiving the microphone 110 , and the laser pointer 220 is disposed in the body 210 . Furthermore, the remote controller 230 can also be disposed in the body 210. In use, the presenter 290 can use the remote control 230 to control the slide 252 played by the computer device 250. Thereby, the projection pen and the long-distance radio microphone are combined with each other, so that the speaker 290 can perform the laser, the slide switch and the sound amplification by holding the pen microphone with one hand when speaking in the slide film. And due to the invention The microphone 110 used is a long-distance radio microphone, and the radio range is wider. Therefore, the microphone is not required to be fixed at a position close to the mouth, and the speaker 290 can freely express the limb according to the content of the speech.

In practice, the shape of the body 210 can be a pen type, a lavalier type, a brooch type or a variety of shapes, and those skilled in the art should flexibly select it at that time.

For example, when the user uses the pen microphone to make a speech, since the pen microphone has the functions of "projection switch", "ray pen" and "microphone", if the user needs to use the microphone during the speech, When the laser pointer is pointed out, the distance between the microphone and the user will change constantly, and this type of use will lead to two problems: one is that the volume and sound will be different depending on the distance of the microphone. The difference is that the composition of the sound band energy will also change due to the difference of the transmission distance. Both problems will cause the discomfort and difficulty of the listener. If the above-mentioned volume control unit 140 is added, The automatic volume adjustment algorithm can automatically detect and adjust the volume of the received voice of the microphone caused by the distance change, and the "dynamic sound equalization algorithm" performed by the sound equalization unit 150 can compare the attenuation of different frequency bands, and Calculate the multiple of the required adjustment of the energy of each frequency band to avoid the situation that the sound frequency band composition changes due to long-distance use, so that Greater flexibility when speaking with a person.

The echo cancellation unit 120, the noise attenuation unit 130, the volume control unit 140, the sound effect equalization unit 150, and the like, as described above, may be a soft body, a hardware body, and/or a carcass body. For example, if the speed of execution Degree and accuracy are the primary considerations, and these units can basically be dominated by hardware and / or carcass; if design flexibility is the primary consideration, then these units can basically be based on software; or, The unit can work together with software, hardware and carcass. It should be understood that the above examples are not intended to limit the present invention, and are not intended to limit the present invention. Those skilled in the art will be able to flexibly select the specific embodiments of the units as needed. For example, the echo cancellation unit 120, the noise attenuation unit 130, the volume control unit 140, and the sound effect equalization unit 150 may be installed in the body 210 or in the computer device 250.

FIG. 3 is a flow diagram of a sound processing method 300 of a microphone system 100 in accordance with an embodiment of the present invention. The sound processing method 300 includes an acoustic echo cancellation step 310, a noise attenuation step 320, an automatic volume adjustment step 330, and a dynamic sound effect equalization step 340 (it is understood that the steps mentioned in this embodiment, except that the order is specifically stated In addition, the order can be adjusted according to actual needs, or even simultaneously or partially.) As for the hardware device for carrying out these steps, since the above embodiments have been specifically disclosed, the description thereof will not be repeated.

In the acoustic echo cancellation step 310, the sound signal broadcast by a speaker is removed from the microphone signal received by a microphone to output a sound wave signal; in the noise attenuation step 320, the microphone is filtered from the sound wave signal. Background noise to output a denoised sound wave signal; in the automatic volume adjustment step 330, comparing the energy similarity between the denoised sound wave signal and a reference model, according to the noise reduction The sound wave signal is adjusted in volume and outputs the sound wave signal after the volume adjustment; In the sound equalization step 340, the degree of similarity between the signal-to-noise frequency energy ratio of the volume-adjusted sound wave signal and the noise-to-noise frequency energy ratio of the reference model is determined according to the volume-adjusted sound wave signal. An audio equalization process is performed and a processed sound wave signal is output to the speaker.

In an embodiment, the acoustic echo cancellation step 310 includes: determining a correlation between the microphone signal and a reference signal, and when the correlation is lower than a preset value, according to a preset offset coefficient. Eliminating the sound signal broadcast by the speaker; otherwise, when the correlation is higher than the preset value, updating the offset coefficient based on the correlation; and eliminating the broadcasted by the speaker according to the updated offset coefficient The sound signal.

Specifically, when the user vocalizes the microphone, the microphone usually receives sound from two sources, one is from the sound of the "current" user, and the voice signal is recorded as "microphone signal"; another voice The sound of the "past" user's voice is played through the speaker. The signal before the "past" speaker is recorded is "reference signal". When performing acoustic echo cancellation, first compare the acoustic characteristics (including spectral data, voiceprint data, amplitude variation, etc.) between the "reference signal" and the "microphone signal". If the two are highly correlated, it means "microphone signal". The composition of most of the components comes from the "reference signal", which can be used to find the filter coefficient between the two, called the "offset factor", by which the offset coefficient will be from the speaker of the "past". The sound is removed, leaving only the sound of the "present" user. Conversely, if the correlation between the two is low, it means that the "microphone signal" component is not only from the "reference signal" in the past, but also contains a considerable proportion of users now making voice messages. At this time, it is not necessary to update the coefficient, but the original "offset factor" is used to deduct the sound from the speaker. After the acoustic echo cancellation processing, the sound signal only has the sound signal from the "current" user. This signal and other acoustic calculations can greatly improve the accuracy of the calculation.

In an embodiment, the noise attenuation step 320 includes: estimating the background noise; and deducting the estimated background noise from the sound wave signal. Specifically, the background signal of the air must be removed before the sound signal is broadcasted by the speaker. Otherwise, the sound signal will be too difficult to be used when the sound is amplified by the speaker. Therefore, the background noise must be performed before the noise deduction calculation is performed. The estimation, after subtracting the estimated background noise from the acoustic echo-removed audio signal, leaves only the sound signal sent by the "current" user without background noise.

In an embodiment, the automatic volume adjustment step 330 includes: determining whether the denoised sound wave signal meets a preset volume condition; and when the denoised sound wave signal meets the preset volume condition, the ratio is decreased. A similarity in energy between the noise signal and the reference model; and volume-adjusting the noise-reduced acoustic signal according to the energy similarity and outputting the volume-adjusted acoustic signal. In an embodiment, the reference model is based on the acoustic characteristics of the white noise collected by the microphone; or, in other embodiments, the reference model is based on various characteristics such as pink noise, gray noise collected by the microphone, and familiar with The artist should consider the need for flexible design at the time.

Specifically, in the automatic volume adjustment step 330, the condition is set to determine whether the residual signal needs to be adjusted in volume; The threshold indicates that the user is close to the microphone, and does not need to perform volume adjustment, or the user is too far away from the microphone to exceed the set microphone range. This situation does not require volume adjustment of the microphone signal. The calculation. If the adjustment threshold is reached, the distance between the user and the microphone is far, but it is still within the range of the microphone. At this time, the volume adjustment is needed to prevent the listener from feeling that the volume is too small and the hearing is unclear. When adjusting the volume, the sound energy at this time must be compared with the sound energy of the pre-recorded standard reference model as the reference basis for adjusting the volume parameter, and then the parameters calculated by the above steps are based on the signals received by the current microphone. Volume adjustment.

Taking the omnidirectional microphone as an example, the threshold of the automatic volume adjustment step 330 is set as follows: if the ratio of the overall acoustic energy to the background noise energy of the current acoustic signal is greater than 2.5 times or less than 0.6 times, it is determined that the current acoustic signal is not required. The automatic volume adjustment process is performed; conversely, if the ratio of the overall sound wave energy to the background noise energy of the current sound wave signal is between 2.5 and 0.6 times, the automatic sound volume adjustment of the current sound wave signal is required to correct the cause and the microphone. The volume changes caused by the distance.

In an embodiment, the dynamic sound equalization step 340 includes: calculating a noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal; determining whether the sound-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to a preset An equalization condition; when the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to the preset equalization condition, the signal-to-noise frequency energy ratio of the volume-adjusted sound wave signal is compared with the reference model The degree of similarity between the noise frequency energy ratios; and the volume adjusted according to the degree of similarity The sound wave signal performs the equalization processing of the sound effect.

Specifically, before the sound effect equalization process, the noise signal received by the microphone must be analyzed for the noise band energy ratio, and the condition is determined to determine whether the sound signal needs to be equalized. If the threshold is not reached, it means that the distance between the user and the microphone is not within the scope of the sound effect. It may be that the user takes the microphone close to the mouth. This use behavior does not cause the sound energy to be attenuated. The listener feels different conditions, so the system determines that there is no need to perform audio equalization processing; or the user is too far away from the microphone to approach the microphone radio range, and also judges that the microphone does not need to receive the sound signal left sound effect, etc. Processing. If the set threshold is reached, the signal-to-noise frequency energy ratio of the sound wave signal is calculated, and the similarity between the signal-to-noise frequency energy ratio of the sound wave signal and the standard reference model is compared with the calculated similarity. As a basis for adjusting the parameters of the sound effect, the sound signal of the residual sound wave signal is equalized.

Taking the omnidirectional microphone as an example, the threshold of the dynamic sound equalization step 340 is set as follows: if the current noise signal frequency band ratio of the sound wave signal is greater than 5 times or less than 0.5 times, it is determined that the current sound wave signal is not required to be sounded. Equivalent processing; conversely, if the current noise signal frequency band of the acoustic signal is between 5 and 0.5 times, it is necessary to perform equalization on some frequency bands of the current acoustic signal to correct the distance from the microphone. The resulting sound quality changes.

Moreover, in one embodiment, the sound processing method 300 includes a reference signal recording step 350 and a sounding step 360. In the reference signal recording step 360, before the processed sound wave signal is sent to the speaker for playback, the processed sound wave signal is recorded first as the acoustic echo cancellation step 310. The "reference signal" used in the next calculation and judgment. Then, in the sounding step 360, the speaker plays the processed sound wave signal.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention, and the present invention can be modified and modified without departing from the spirit and scope of the present invention. The scope is subject to the definition of the scope of the patent application attached.

100‧‧‧Microphone system

110‧‧‧ microphone

120‧‧‧Echo cancellation unit

130‧‧‧Noise attenuation unit

140‧‧‧Volume control unit

150‧‧‧Sound equalization unit

190‧‧‧Speaker

Claims (16)

A microphone system includes: a microphone; an echo cancellation unit configured to remove an audio signal broadcast by a speaker from the microphone signal received by the microphone to output a sound wave signal; a noise attenuation unit for self The sound signal of the microphone is filtered out to output a denoised sound wave signal; and a volume control unit is used to compare the energy similarity between the denoised sound wave signal and a reference model. Corresponding to the volume adjustment of the noise-reduced acoustic wave signal and outputting the volume-adjusted sound wave signal; and an audio equalization unit for comparing the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal with Referring to the degree of similarity between the signal-to-noise frequency energy ratio of the reference model, an acoustic equalization process is performed on the volume-adjusted sound wave signal and a processed sound wave signal is output to the speaker. The microphone system of claim 1, wherein the echo cancellation unit determines a correlation between the microphone signal and a reference signal, and when the correlation is lower than a preset value, the echo cancellation unit is configured according to one of the presets The offset factor is used to eliminate the sound signal broadcast by the speaker. The microphone system of claim 2, wherein when the correlation is higher than the preset value, the echo cancellation unit updates the offset based on the correlation The sales factor eliminates the sound signal broadcast by the speaker based on the updated offset factor. The microphone system of claim 2, wherein the sound effect equalization unit calculates a signal-to-noise frequency energy ratio of the volume-adjusted sound wave signal, and determines whether the sound-to-noise frequency energy ratio of the volume-adjusted sound wave signal matches a preset equalization condition, when the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to the preset equalization condition, the sound effect equalization unit compares the noise noise frequency of the volume-adjusted sound wave signal The degree of similarity between the energy ratio and the noise-to-noise frequency energy ratio of the reference model, and performing the sound equalization process on the volume-adjusted sound wave signal according to the degree of similarity. The microphone system of claim 4, wherein the sound effect equalization unit records the processed sound wave signal as the reference signal before sending the processed sound wave signal to the speaker for playing. The microphone system of claim 1, wherein the reference model is based on acoustic characteristics of white noise collected by the microphone. The microphone system of claim 1, wherein the volume control unit determines whether the denoised sound wave signal meets a preset volume condition, and when the noise-reduced sound wave signal meets the preset volume condition, the volume control The unit ratio is similar to the energy between the denoised sound wave signal and the reference model, and the noise is reduced according to the energy similarity The wave signal makes a volume adjustment and outputs the volume-adjusted sound wave signal. The microphone system of claim 1, wherein the noise attenuation unit subtracts the estimated background noise from the sound wave signal after estimating the background noise. A sound processing method is applicable to a microphone system, and the sound processing method comprises the following steps: (a) eliminating an audio signal broadcast by a speaker from a microphone signal received by a microphone to output a sound wave signal; (b) Filtering the background noise of the microphone from the sound wave signal to output a noise-reduced acoustic wave signal; (c) comparing the energy similarity between the noise-reduced acoustic wave signal and a reference model, according to The noise-reduced acoustic wave signal is volume-adjusted and outputs a volume-adjusted sound wave signal; and (d) a ratio of the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal to the noise-to-noise frequency energy ratio of the reference model The degree of similarity between them is such that an audio effect equalization process is performed on the volume-adjusted sound wave signal and a processed sound wave signal is output to the speaker. The sound processing method of claim 9, wherein the step (a) comprises: determining a correlation between the microphone signal and a reference signal; and when the correlation is lower than a preset value, according to one of the presets Offset coefficient To eliminate the sound signal broadcast by the speaker. The sound processing method of claim 10, wherein the step (a) further comprises: when the correlation is higher than the preset value, updating the offset coefficient based on the correlation; and the offsetting according to the update The coefficient is used to eliminate the sound signal broadcast by the speaker. The sound processing method of claim 10, wherein the step (d) comprises: calculating a noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal; determining whether the sound-frequency-frequency energy ratio of the volume-adjusted sound wave signal is Corresponding to a preset equalization condition; when the noise-to-noise frequency energy ratio of the volume-adjusted sound wave signal conforms to the preset equalization condition, the signal-to-noise frequency energy ratio of the volume-adjusted sound wave signal is compared with the The degree of similarity between the signal-to-noise frequency energy ratio of the reference model; and performing the sound equalization process on the volume-adjusted sound wave signal according to the degree of similarity. The sound processing method of claim 12, wherein the step (d) further comprises: before sending the processed sound wave signal to the speaker for playing, The processed sound wave signal is recorded as the reference signal. The sound processing method of claim 9, wherein the reference model is based on acoustic characteristics of white noise collected by the microphone. The sound processing method of claim 9, wherein the step (c) comprises: determining whether the noise-reduced sound wave signal meets a preset volume condition; and when the noise-reduced sound wave signal meets the preset volume condition, Comparing the energy similarity between the noise-reduced acoustic wave signal and the reference model; and adjusting the noise-reduced acoustic wave signal according to the energy magnitude similarity and outputting the volume-adjusted acoustic signal . The sound processing method of claim 9, wherein the step (b) comprises: estimating the background noise; and deducting the estimated background noise from the sound wave signal.