CN103812462B - Volume control method and device - Google Patents
Volume control method and device Download PDFInfo
- Publication number
- CN103812462B CN103812462B CN201210460201.0A CN201210460201A CN103812462B CN 103812462 B CN103812462 B CN 103812462B CN 201210460201 A CN201210460201 A CN 201210460201A CN 103812462 B CN103812462 B CN 103812462B
- Authority
- CN
- China
- Prior art keywords
- gain
- signal
- loudness
- level
- adjustment
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Landscapes
- Circuit For Audible Band Transducer (AREA)
- Tone Control, Compression And Expansion, Limiting Amplitude (AREA)
Abstract
The embodiment of the present invention provides a kind of volume control method and device, and this volume control method includes: at least one road acoustical signal is carried out speech detection respectively；For each road acoustical signal, according to the voice signal detected, determine the speech loudness gain of voice signal, according to the mute signal detected, determine the gain of the largest tolerable of mute signal；Gain generating according to speech loudness gain and largest tolerable adjusts gain；According to adjusting gain, acoustical signal is carried out Gain tuning.The volume control method of embodiment of the present invention offer and device, it is achieved that under the application scenarios that noise is more or during real-time Communication for Power, the loudness to audio frequency controls, and improves the effect that loudness controls.
Description
Technical field
The present embodiments relate to audio signal processing technique, particularly relate to a kind of volume control method and device.
Background technology
Loudness is the index that the signal of telecommunication is converted to voice metric energy during acoustic vibration, is human auditory system
Subjective feeling to sound intensity.Loudness controls (Loudness Control) mainly according to the loudness of signal
Different to different frequency range perception with human ear, increase or deamplification so that acoustic playback is maintained at one
In identical perception level or identical loudness.
The loudness of prior art controls to be mainly used in audio player, the audio frequency of audio player plays
Voice quality generally relatively good.But, under the application scenarios that noise is more or real-time Communication for Power
During, control to carry out Gain tuning by the loudness of prior art, can be by noise according to the adjustment of voice
Gain is adjusted, poor effect.
Summary of the invention
The embodiment of the present invention provides a kind of volume control method and device, to realize in the more application of noise
Loudness to audio frequency controls under scene or during real-time Communication for Power, improves the effect that loudness controls.
First aspect, the embodiment of the present invention provides a kind of volume control method, including:
At least one road acoustical signal is carried out speech detection respectively；
For each road acoustical signal, according to the voice signal detected, determine the language of described voice signal
Sonority gain, according to the mute signal detected, determines the increasing of the largest tolerable of described mute signal
Benefit；
Gain generating according to described speech loudness gain and described largest tolerable adjusts gain；
According to described adjustment gain, described acoustical signal is carried out Gain tuning.
In the implementation that the first is possible, described at least one road acoustical signal is carried out voice inspection respectively
Survey, including:
For described each road acoustical signal, calculate the root-mean-square of described acoustical signal；
Signal envelope and noise envelope is generated respectively according to described root-mean-square；
Calculate described signal envelope and the ratio of described noise envelope, if described ratio presets threshold more than first
Value, then detect described voice signal, otherwise, described mute signal detected.
In conjunction with the first possible implementation of first aspect, in the implementation that the second is possible,
The mute signal that described basis detects, determines the gain of the largest tolerable of described mute signal, specifically
For:
According to the level of mute signal described in described root mean square calculation, the level according to described mute signal is true
The gain of fixed described largest tolerable.
In conjunction with the implementation that the second of first aspect is possible, in the implementation that the third is possible,
The described level according to mute signal described in described root mean square calculation, particularly as follows:
The level Noise_Level of the application below equation described mute signal of calculating:
Noise_Level=0.99×Noise_Level+0.01×Ecur；
Wherein, Ecur is described root-mean-square.
In the 4th kind of possible implementation, the voice signal that described basis detects, determine institute's predicate
The speech loudness gain of tone signal, particularly as follows:
Described voice signal is carried out loudness Filtering Processing, the voice signal after loudness Filtering Processing is carried out
Signal level is added up, and determines the level of described voice signal according to statistical result, according to described voice signal
Level determine described speech loudness gain.
In the 5th kind of possible implementation, described can according to described speech loudness gain and described maximum
The Gain generating of tolerance adjusts gain, particularly as follows:
If the absolute value of described speech loudness gain is more than the second predetermined threshold value, then application below equation generates
Described adjustment gain G ain:
Gain=LGain×(1.0-(LGain+NGain)/(LGain×2))；
Wherein, LGain is described speech loudness gain, and NGain is the gain of described largest tolerable.
In the 6th kind of possible implementation, described according to described adjustment gain, described acoustical signal is entered
Row Gain tuning, particularly as follows:
Adjustment duration is determined according to described acoustical signal, true according to described adjustment gain and described adjustment duration
Synchronizing of setting the tone is long, according to described adjustment gain and described adjustment step-length, described acoustical signal is carried out gain tune
Whole.
In the 7th kind of possible implementation, if described acoustical signal is at least two-way, the most described basis
The Gain generating of described speech loudness gain and described largest tolerable adjusts after increasing, described in described basis
Before adjustment gain carries out Gain tuning to described acoustical signal, described method also includes:
Calculate described speech loudness gain and the difference of described adjustment gain on each road, determine in each road
Big difference, described adjustment gain G ain on each road is adjusted by application below equation:
Gain=2×Gain–LDiffMax–LGain；
Wherein, LDiffMax is the difference of described maximum, and LGain is described speech loudness gain.
Second aspect, the embodiment of the present invention provides a kind of loudness to control device, including:
Detector unit, for carrying out speech detection respectively at least one road acoustical signal；
First processing unit, is connected with described detector unit, for for each road acoustical signal, according to
The voice signal detected, determines the speech loudness gain of described voice signal, quiet according to detect
Signal, determines the gain of the largest tolerable of described mute signal；
Second processing unit, is connected with described first processing unit, for according to described speech loudness gain
Gain is adjusted with the Gain generating of described largest tolerable；
First adjustment unit, is connected with described second processing unit, is used for according to described adjustment gain institute
State acoustical signal and carry out Gain tuning.
In the implementation that the first is possible, described detector unit includes:
First processes subelement, for for described each road acoustical signal, calculates described acoustical signal
Root-mean-square；
Second processes subelement, processes subelement with described first and is connected, for dividing according to described root-mean-square
Sheng Cheng signal envelope and noise envelope；
Judgment sub-unit, processes subelement with described second and is connected, be used for calculating described signal envelope and institute
State the ratio of noise envelope, if described ratio is more than the first predetermined threshold value, then described voice signal detected,
Otherwise, described mute signal is detected.
In conjunction with the first possible implementation of second aspect, in the implementation that the second is possible,
Described first processing unit specifically for the level according to mute signal described in described root mean square calculation, according to
The level of described mute signal determines the gain of described largest tolerable.
In conjunction with the implementation that the second of second aspect is possible, in the implementation that the third is possible,
Described first processing unit calculates the level of described mute signal specifically for application below equation
Noise_Level:
Noise_Level=0.99×Noise_Level+0.01×Ecur；
Wherein, Ecur is described root-mean-square.
In the 4th kind of possible implementation, described first processing unit is specifically for believing described voice
Number carry out loudness Filtering Processing, the voice signal after loudness Filtering Processing is carried out signal level statistics, root
Result determines the level of described voice signal according to statistics, determines institute's predicate according to the level of described voice signal
Sonority gain.
In the 5th kind of possible implementation, if described second processing unit is specifically for institute's Voice Response
The absolute value of degree gain is more than the second predetermined threshold value, then application below equation generates described adjustment gain G ain:
Gain=LGain×(1.0-(LGain+NGain)/(LGain×2))；
Wherein, LGain is described speech loudness gain, and NGain is the gain of described largest tolerable.
In the 6th kind of possible implementation, described first adjustment unit is specifically for according to described sound
Signal determines adjustment duration, determines adjustment step-length according to described adjustment gain and described adjustment duration, according to
Described adjustment gain and described adjustment step-length carry out Gain tuning to described acoustical signal.
In the 7th kind of possible implementation, if described acoustical signal is at least two-way, described loudness control
Device processed also includes:
Second adjustment unit, is connected with described first adjustment unit, for calculating the described voice on each road
Loudness gain and the difference of described adjustment gain, determine difference maximum in each road, applies below equation pair
Described adjustment gain G ain on each road is adjusted:
Gain=2×Gain-LDiffMax-LGain；
Wherein, LDiffMax is the difference of described maximum, and LGain is described speech loudness gain.
As shown from the above technical solution, the embodiment of the present invention provide volume control method and device, loudness
Control device and at least one road acoustical signal is carried out speech detection respectively, for each road acoustical signal, root
According to the voice signal detected, determine the speech loudness gain of voice signal, according to the quiet letter detected
Number, determine the gain of the largest tolerable of mute signal, according to speech loudness gain and largest tolerable
Gain generating adjusts gain, according to adjusting gain, acoustical signal is carried out Gain tuning.By to quiet letter
Number detection determine the gain of largest tolerable, and simultaneously according to speech loudness gain and largest tolerable
Gain generating adjusts gain, is adjusted the gain of acoustical signal by this adjustment gain so that adjust
After acoustical signal more conform to the perception level of human ear, greatly increase the effect that loudness controls.
Accompanying drawing explanation
In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to reality
Execute the required accompanying drawing used in example or description of the prior art to be briefly described, it should be apparent that under,
Accompanying drawing during face describes is some embodiments of the present invention, for those of ordinary skill in the art,
On the premise of not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
The first volume control method flow chart that Fig. 1 provides for the embodiment of the present invention；
The second volume control method flow chart that Fig. 2 provides for the embodiment of the present invention；
The first loudness that Fig. 3 provides for the embodiment of the present invention controls apparatus structure schematic diagram；
The second loudness that Fig. 4 provides for the embodiment of the present invention controls apparatus structure schematic diagram；
The third loudness that Fig. 5 provides for the embodiment of the present invention controls apparatus structure schematic diagram.
Detailed description of the invention
For making the purpose of the embodiment of the present invention, technical scheme and advantage clearer, below in conjunction with this
Accompanying drawing in bright embodiment, is clearly and completely described the technical scheme in the embodiment of the present invention,
Obviously, described embodiment is a part of embodiment of the present invention rather than whole embodiments.Based on
Embodiment in the present invention, those of ordinary skill in the art are obtained under not making creative work premise
The every other embodiment obtained, broadly falls into the scope of protection of the invention.
The first volume control method flow chart that Fig. 1 provides for the embodiment of the present invention.As it is shown in figure 1,
The volume control method that the present embodiment provides specifically can apply to but is not limited under virtual meeting scene sound
The process that controls in real time of the loudness of tone signal, the volume control method that the present embodiment provides can pass through loudness
Controlling device to perform, this loudness controls device and can be integrated in audio processing equipment, it is also possible to individually
Arrange.This audio processing equipment specifically can apply in conference system.
The volume control method that the present embodiment provides specifically includes:
Step 10, at least one road acoustical signal is carried out speech detection respectively；
Step 20, for each road acoustical signal, according to the voice signal detected, determine described voice
The speech loudness gain of signal, according to the mute signal detected, determines that the maximum of described mute signal can
The gain of tolerance；
Step 30, adjust gain according to the Gain generating of described speech loudness gain and described largest tolerable；
Step 40, according to described adjustment gain, described acoustical signal is carried out Gain tuning.
Specifically, at least one road acoustical signal is the signal needing to carry out loudness control, when in virtual meeting
Under view application scenarios, acoustical signal is usually multichannel.Audio processing equipment receives the monophonic of each terminal
Code stream, obtain a road acoustical signal after being decoded a monaural code stream processing, more right
Each road acoustical signal carries out loudness control.
The acoustical signal received is probably voice signal and is also likely to be mute signal, when a user speaks,
This acoustical signal is voice signal, and when user does not speak, this acoustical signal is mute signal.To sound
Tone signal carries out speech detection, voice signal or mute signal during to judge this acoustical signal.This voice
Detection process specifically can use Voice activity detector (Voice Active Detection is called for short VAD)
Method realizes,
When voice signal being detected, determine the speech loudness gain of voice signal, when quiet letter being detected
Number time, information entrained in mute signal is usually noise, determines the largest tolerable of mute signal
Gain, the calculating of the gain of speech loudness gain and largest tolerable specifically can be according to playback gain
(Replay Gain) standard realizes.Gain generating according to speech loudness gain and largest tolerable is adjusted
Whole gain, is adjusted the gain of this road acoustical signal by adjusting gain, to realize loudness control.
The volume control method that the present embodiment provides, loudness controls device at least one road acoustical signal respectively
Carry out speech detection, for each road acoustical signal, according to the voice signal detected, determine that voice is believed
Number speech loudness gain, according to the mute signal detected, determine the largest tolerable of mute signal
Gain, adjusts gain according to the Gain generating of speech loudness gain and largest tolerable, according to adjusting gain
Acoustical signal is carried out Gain tuning.By the detection of mute signal being determined the gain of largest tolerable,
And adjust gain according to the Gain generating of speech loudness gain and largest tolerable simultaneously, increased by this adjustment
The gain of acoustical signal is adjusted by benefit so that the acoustical signal after adjustment more conforms to the perception of human ear
Level, greatly increases the effect that loudness controls.
The second volume control method flow chart that Fig. 2 provides for the embodiment of the present invention.As in figure 2 it is shown,
In the present embodiment, step 10, described at least one road acoustical signal is carried out speech detection respectively, specifically
May include steps of:
Step 101, for described each road acoustical signal, calculate the root-mean-square of described acoustical signal；
Step 102, generate signal envelope and noise envelope respectively according to described root-mean-square；
Step 103, calculate described signal envelope and the ratio of described noise envelope, if described ratio is more than
First predetermined threshold value, then detect described voice signal, otherwise, described mute signal detected.
Specifically, acoustical signal is digital signal, may determine that sound by calculating the root-mean-square of acoustical signal
The energy of tone signal.Acoustical signal includes that multiple audio frame sequence, each audio frame include multiple sampled point,
The detection process of acoustical signal is the detection process to audio frame.Such as, audio frame is s (n), n=0,
1 ..., N-1, N are the number of sampled point, when the frame length of s (n) is 20ms, can pass through below equation
Calculate the root-mean-square Ecur of s (n).
The process generating signal envelope Senv according to root-mean-square Ecur is specifically as follows:
If Ecur is more than thread1, then Senv=0.9 × Senv+0.1 × Ecur；
If Ecur is not more than thread1, then Senv=0.998 × Senv+0.002 × Ecur；
The process generating noise envelope Sno according to root-mean-square Ecur is specifically as follows:
If Ecur is more than thread2, then Sno=0.998 × Sno+0.002 × Ecur；
If Ecur is not more than thread2, then Sno=0.9 × Sno+0.1 × Ecur；
The initial value of thread1 and thread2 can be arranged based on experience value, and thread1 >=thread2,
During processing, further according to Senv and Sno, thread1 and thread2 is adjusted.As, thread2
=(Senv+Sno) × 0.5, thread1 could be arranged to equal with thread2, or slightly larger than thread2.
The ratio Senv/Sno of signal calculated envelope and noise envelope, it is judged that whether Senv/Sno is more than
Thread3, the most then this acoustical signal is voice signal, and otherwise, this acoustical signal is noise signal.
It should be noted that the parameter in above-mentioned formula all can arrange according to actual treatment situation and adjust
Whole.
In the present embodiment, in step 20, the mute signal that described basis detects, determine described quiet
The gain of the largest tolerable of signal, is specifically as follows:
According to the level of mute signal described in described root mean square calculation, the level according to described mute signal is true
The gain of fixed described largest tolerable.
Specifically, can first determine the maximum noise level that human ear can be tolerated, for example, 30dB, then
Difference according to this maximum noise level Yu the level of mute signal determines the gain of this largest tolerable, i.e.
In the range of human ear can be tolerated, mute signal is adjusted.
In the present embodiment, the described level according to mute signal described in described root mean square calculation, specifically may be used
Think:
The level Noise_Level of the application below equation described mute signal of calculating:
Noise_Level=0.99×Noise_Level+0.01×Ecur；
Wherein, Ecur is described root-mean-square.
Assume that the maximum noise level that human ear can be tolerated is NoiseThread, then the gain of largest tolerable
NGain=NoiseThread-Noise_Level。
In the present embodiment, in step 20, the voice signal that described basis detects, determine described voice
The speech loudness gain of signal, is specifically as follows:
Described voice signal is carried out loudness Filtering Processing, the voice signal after loudness Filtering Processing is carried out
Signal level is added up, and determines the level of described voice signal according to statistical result, according to described voice signal
Level determine described speech loudness gain.
Specifically, can use loudness wave filter that voice signal is carried out loudness Filtering Processing, this loudness is filtered
Ripple device can be with the high pass IIR yulewalk wave filter on 10 rank and the Butterworth high pass of 2 order frequency 150HZ
Wave filter cascade approaches this loudness wave filter.High pass IIR yulewalk wave filter and Butterworth high pass
The parameter of wave filter can need to arrange according to actual process, is not limited with the present embodiment.
Voice signal after loudness Filtering Processing carries out signal level statistics be specifically as follows: calculate voice
The root-mean-square Level of each audio frame in signal, audio frame is s (n), n=0,1 ..., N-1, N are for adopting
The number of sampling point, when the frame length of s (n) is 20ms, and sample rate is 16Hz, root-mean-square Level specifically may be used
To be realized by below equation:
The root-mean-square of multiple audio frames is carried out level statistics, if the level distribution of major part audio frame, certain
Around individual level, then can determine the level Level1 of voice signal according to this level.Can be by following
Formula calculating speech loudness gain LGain:
LGain=Lref–Level1；Wherein, Lref is specifically as follows an empirical value, for example,-14dB.
In the present embodiment, step 30, described according to described speech loudness gain and described largest tolerable
Gain generating adjust gain, be specifically as follows:
If the absolute value of described speech loudness gain is more than the second predetermined threshold value, then application below equation generates
Described adjustment gain G ain:
Gain=LGain×(1.0-(LGain+NGain)/(LGain×2))；
Wherein, LGain is described speech loudness gain, and NGain is the gain of described largest tolerable.
Specifically, first absolute value and second predetermined threshold value of speech loudness gain being compared, this is years old
Two predetermined threshold value are specifically as follows an empirical value, for example, 3dB..Absolute value when speech loudness gain
During more than the second predetermined threshold value, being generated by above-mentioned formula and adjust gain, the parameter in above-mentioned formula also may be used
To arrange according to actual needs, the present embodiment merely provides one preferred embodiment, the present invention
It is not limited thereto.
In the present embodiment, step 40, described according to described adjustment gain, described acoustical signal is increased
Benefit adjusts, particularly as follows:
Determine adjustment duration according to described acoustical signal, determine according to described adjustment gain and described adjustment duration
Adjust step-length, according to described adjustment gain and described adjustment step-length, described acoustical signal is carried out Gain tuning.
Acoustical signal is carried out Gain tuning and can use automatic growth control (Automatic Gain
Control, is called for short AGC) method, detailed process is:
Signal characteristic according to acoustical signal calculates and adjusts step-length, first can be according to the signal of acoustical signal
Type calculates and adjusts duration decay, such as: decay=Ratio × MaxFrameNum × FrameLen+
FrameLen, wherein, Ratio is acoustical signal similarity when being to speak state, MaxFrameNum
For largest frames long number, FrameLen is the length of every frame,.
Calculate adjustment step-length delt:delt=(the curGain-m_oldGain)/decay of each sampled point,
CurGain is the gain of current sampling point, and m_oldGain is the gain of last sampled point, for first
The gain of sampled point could be arranged to 1.
Each audio frame of acoustical signal is carried out Gain tuning, S ' (n)=S (n) × (m_oldGain+delt)
And update oldGain.This adjustment process is specifically as follows:
The gain of last sampled point is m_oldGain, and curGain is the current sampling point after some frames
Gain, decay is the duration of some frames.
The first step, application formula delt=(curGain-m_oldGain)/decay, calculate a pointwise
The delt updated.
Second step, updates each sampled point:
m_curGain=m_oldGain+delt；
S’(n)=S(n)×m_curGain；
m_oldGain=m_curGain；
Here adding the gain of a current sample, above iteration is until this frame signal has processed.
After coming Deng lower frame signal, can be according to the processing procedure more than characteristic cycle of new signal.
In the present embodiment, if described acoustical signal is at least two-way, then step 30, described in described basis
The Gain generating of speech loudness gain and described largest tolerable adjusts after increasing, step 40, described basis
Before described adjustment gain carries out Gain tuning to described acoustical signal, described method can also include:
Step 50, the described speech loudness gain calculating each road and the difference of described adjustment gain, determine
Difference maximum in each road, described adjustment gain G ain on each road is adjusted by application below equation:
Gain=2×Gain–LDiffMax–LGain；
Wherein, LDiffMax is the difference of described maximum, and LGain is described speech loudness gain.
Specifically, under virtual meeting application scenarios, it is usually present multi-channel sound signal, for example, M road,
The adjustment gain of m road acoustical signal is Gain (m), [m=1..M].Ring at the voice calculating each road
During degree gain, it is identical with reference to Lref, if the most each Lu Douyong LGain (m) is carried out
Adjust, then the energy after adjustment is Lref, when the noise level difference on every road when, finally calculates
Obtain Gain (m), each road acoustical signal through Gain (m) adjust after energy be different, by with
Under type adjusts the adjustment energy of every road acoustical signal that aligns automatically so that the level of output is identical.
First the acoustical signal difference to Lref on every road is calculated: LDiff (m)=LGain (m)-Gain (m), from
In choose maximum LDiffMax, i.e. LDiffMax=Max (LDiff (m)), then by formula Gain=2
Described adjustment gain G ain on each road is adjusted by × Gain LDiffMax LGain.
In actual application, can also be according to presetting through Gain tuning Hou Ge road acoustical signal
Scene add azimuth information to every road acoustical signal, synthesize 3d audio frequency.
The first loudness that Fig. 3 provides for the embodiment of the present invention controls apparatus structure schematic diagram.Such as Fig. 3 institute
Showing, the loudness that the present embodiment provides controls device specifically can realize the sound that any embodiment of the present invention provides
Each step of degree control method, the process of implementing does not repeats them here.The loudness that the present embodiment provides
Control device can be integrated in audio processing equipment, it is also possible to be separately provided.This audio processing equipment has
Body can apply in conference system.
The present embodiment provide loudness control device specifically include detector unit the 11, first processing unit 12,
Second processing unit 13 and the first adjustment unit 14.Described detector unit 11 is at least one road sound
Signal carries out speech detection respectively.Described first processing unit 12 is connected with described detector unit 11, uses
In for each road acoustical signal, according to the voice signal detected, determine the voice of described voice signal
Loudness gain, according to the mute signal detected, determines the gain of the largest tolerable of described mute signal.
Described second processing unit 13 is connected with described first processing unit 12, for according to described speech loudness
The Gain generating of gain and described largest tolerable adjusts gain.Described first adjustment unit 14 and described the
Two processing units 13 are connected, for described acoustical signal being carried out Gain tuning according to described adjustment gain.
The loudness that the present embodiment provides controls device, and at least one road acoustical signal is entered by detector unit 11 respectively
Row speech detection, the first processing unit 12 is for each road acoustical signal, according to the voice signal detected,
Determine the speech loudness gain of voice signal, according to the mute signal detected, determine that mute signal is
Big tolerable gain, the second processing unit 13 is raw according to the gain of speech loudness gain and largest tolerable
Becoming to adjust gain, the first adjustment unit 14 carries out Gain tuning according to adjusting gain to acoustical signal.Pass through
Detection to mute signal determines the gain of largest tolerable, and simultaneously according to speech loudness gain and maximum
Tolerable Gain generating adjusts gain, is adjusted the gain of acoustical signal by this adjustment gain,
Make the acoustical signal after adjusting more conform to the perception level of human ear, greatly increase what loudness controlled
Effect.
The second loudness that Fig. 4 provides for the embodiment of the present invention controls apparatus structure schematic diagram.Such as Fig. 4 institute
Show, in the present embodiment, described detector unit 11 specifically can include the first process subelement 21,
Two process subelement 22 and judgment sub-unit 23.Described first process subelement 21 for for described often
One road acoustical signal, calculates the root-mean-square of described acoustical signal.Described second processes subelement 22 with described
First processes subelement 21, for generating signal envelope and noise envelope respectively according to described root-mean-square.Institute
State judgment sub-unit 23 to process subelement 22 with described second and be connected, be used for calculating described signal envelope with
The ratio of described noise envelope, if described ratio is more than the first predetermined threshold value, then detects that described voice is believed
Number, otherwise, described mute signal detected.
Specifically, acoustical signal is digital signal, may determine that sound by calculating the root-mean-square of acoustical signal
The energy of tone signal.Acoustical signal includes that multiple audio frame sequence, each audio frame include multiple sampled point,
The detection process of acoustical signal is the detection process to audio frame.Such as, audio frame is s (n), n=0,
1 ..., N-1, N are the number of sampled point, when the frame length of s (n) is 20ms, can pass through below equation
Calculate the root-mean-square Ecur of s (n).
The process generating signal envelope Senv according to root-mean-square Ecur is specifically as follows:
If Ecur is more than thread1, then Senv=0.9 × Senv+0.1 × Ecur；
If Ecur is not more than thread1, then Senv=0.998 × Senv+0.002 × Ecur；
The process generating noise envelope Sno according to root-mean-square Ecur is specifically as follows:
If Ecur is more than thread2, then Sno=0.998 × Sno+0.002 × Ecur；
If Ecur is not more than thread2, then Sno=0.9 × Sno+0.1 × Ecur；
The initial value of thread1 and thread2 can be arranged based on experience value, and thread1 >=thread2,
During processing, further according to Senv and Sno, thread1 and thread2 is adjusted.As, thread2
=(Senv+Sno) × 0.5, thread1 could be arranged to equal with thread2, or slightly larger than thread2.
The ratio Senv/Sno of signal calculated envelope and noise envelope, it is judged that whether Senv/Sno is more than
Thread3, the most then this acoustical signal is voice signal, and otherwise, this acoustical signal is noise signal.
It should be noted that the parameter in above-mentioned formula all can arrange according to actual treatment situation and adjust
Whole.
In the present embodiment, described first processing unit 12 specifically may be used for according to described root mean square calculation
The level of described mute signal, determines the gain of described largest tolerable according to the level of described mute signal.
In the present embodiment, described first processing unit 12 specifically can be specifically for application below equation meter
Calculate the level Noise_Level of described mute signal:
Noise_Level=0.99×Noise_Level+0.01×Ecur；
Wherein, Ecur is described root-mean-square.
In the present embodiment, described first processing unit 12 specifically may be used for ringing described voice signal
Degree Filtering Processing, carries out signal level statistics to the voice signal after loudness Filtering Processing, according to statistical result
Determine the level of described voice signal, determine described speech loudness gain according to the level of described voice signal.
Specifically, the first processing unit 12 can arrange loudness wave filter, by loudness wave filter to language
Tone signal carries out loudness Filtering Processing, and this loudness wave filter can filter with the high pass IIR yulewalk on 10 rank
The butterworth high pass filter cascade of device and 2 order frequency 150HZ approaches this loudness wave filter.High pass
The parameter of IIR yulewalk wave filter and butterworth high pass filter can be according to actual process need
Arrange, be not limited with the present embodiment.
In the present embodiment, if exhausted specifically for described speech loudness gain of described second processing unit 13
To value more than the second predetermined threshold value, then described adjustment gain G ain of application below equation generation:
Gain=LGain×(1.0-(LGain+NGain)/(LGain×2))；
Wherein, LGain is described speech loudness gain, and NGain is the gain of described largest tolerable.
In the present embodiment, described first adjustment unit 14 is specifically for determining tune according to described acoustical signal
Whole duration, determines adjustment step-length according to described adjustment gain and described adjustment duration, increases according to described adjustment
Beneficial and described adjustment step-length carries out Gain tuning to described acoustical signal.
In the present embodiment, further, if described acoustical signal is at least two-way, the control of described loudness
Device can also include that the second adjustment unit 15, described second adjustment unit 15 adjust list with described first
Unit 14 is connected, for calculating described speech loudness gain and the difference of described adjustment gain on each road, really
Difference maximum in Ding Ge road, described adjustment gain G ain on each road is adjusted by application below equation:
Gain=2×Gain-LDiffMax-LGain；
Wherein, LDiffMax is the difference of described maximum, and LGain is described speech loudness gain.
Under virtual meeting application scenarios, it is usually present multi-channel sound signal, for example, M road, m road
The adjustment gain of acoustical signal is Gain (m), [m=1..M].Speech loudness gain on each road of calculating
During, it is identical with reference to Lref, if the most each Lu Douyong LGain (m) is adjusted, that
Energy after Tiao Zhenging is Lref, when the noise level difference on every road when, is finally calculated
Gain (m), acoustical signal energy after Gain (m) adjusts in each road is different, by with lower section
Formula adjusts the adjustment energy of every road acoustical signal that aligns automatically so that the level of output is identical.
First the acoustical signal difference to Lref on every road is calculated: LDiff (m)=LGain (m)-Gain (m), from
In choose maximum LDiffMax, i.e. LDiffMax=Max (LDiff (m)), then by formula Gain=2
Described adjustment gain G ain on each road is adjusted by × Gain LDiffMax LGain.
In actual application, can also be according to presetting through Gain tuning Hou Ge road acoustical signal
Scene add azimuth information to every road acoustical signal, synthesize 3d audio frequency.
The third loudness that Fig. 5 provides for the embodiment of the present invention controls apparatus structure schematic diagram.Such as Fig. 5 institute
Showing, the loudness that the present embodiment provides controls device specifically can realize the sound that any embodiment of the present invention provides
Each step of degree control method, the process of implementing does not repeats them here.The loudness that the present embodiment provides
Controlling device and specifically include processor 31 and memorizer 32, described memorizer 32 is used for storing instruction.Institute
State processor 31 to couple with described memorizer 32, described processor 31 be configured to execution be stored in described in deposit
Instruction in reservoir 32, wherein, described processor 31 is configurable at least one road acoustical signal
Carry out speech detection respectively, for each road acoustical signal, according to the voice signal detected, determine institute
The speech loudness gain of predicate tone signal, according to the mute signal detected, determines described mute signal
The gain of largest tolerable, adjusts according to the Gain generating of described speech loudness gain and described largest tolerable
Whole gain, carries out Gain tuning according to described adjustment gain to described acoustical signal.
One of ordinary skill in the art will appreciate that: all or part of step realizing said method embodiment can
Completing with the hardware relevant by programmed instruction, aforesaid program can be stored in an embodied on computer readable and deposit
In storage media, this program upon execution, performs to include the step of said method embodiment；And aforesaid storage
Medium includes: the various media that can store program code such as ROM, RAM, magnetic disc or CD.
Last it is noted that above example is only in order to illustrate technical scheme, rather than to it
Limit；Although the present invention being described in detail with reference to previous embodiment, the ordinary skill of this area
Personnel it is understood that the technical scheme described in foregoing embodiments still can be modified by it, or
Person carries out equivalent to wherein portion of techniques feature；And these amendments or replacement, do not make corresponding skill
The essence of art scheme departs from the scope of various embodiments of the present invention technical scheme.
Claims (14)
1. a volume control method, it is characterised in that including:
At least one road acoustical signal is carried out speech detection respectively；
For each road acoustical signal, according to the voice signal detected, determine the voice of described voice signal
Loudness gain, according to the mute signal detected, determines the gain of the largest tolerable of described mute signal；
Gain generating according to described speech loudness gain and described largest tolerable adjusts gain；
According to described adjustment gain, described acoustical signal is carried out Gain tuning；
Wherein, described at least one road acoustical signal is carried out speech detection respectively, including:
For described each road acoustical signal, calculate the root-mean-square of described acoustical signal；
Signal envelope and noise envelope is generated respectively according to described root-mean-square；
Calculate described signal envelope and the ratio of described noise envelope, if described ratio presets threshold more than first
Value, then detect described voice signal, otherwise, described mute signal detected.
Volume control method the most according to claim 1, it is characterised in that described basis detects
Mute signal, determine the gain of the largest tolerable of described mute signal, particularly as follows:
According to the level of mute signal described in described root mean square calculation, the level according to described mute signal is true
The gain of fixed described largest tolerable.
Volume control method the most according to claim 2, it is characterised in that described according to described all
Root calculates the level of described mute signal, particularly as follows:
The level Noise_Level of the application below equation described mute signal of calculating:
Noise_Level=0.99 × Noise_Level+0.01 × Ecur；
Wherein, Ecur is described root-mean-square.
Volume control method the most according to claim 1, it is characterised in that described basis detects
Voice signal, determine the speech loudness gain of described voice signal, particularly as follows:
Described voice signal is carried out loudness Filtering Processing, the voice signal after loudness Filtering Processing is carried out
Signal level is added up, and determines the level of described voice signal according to statistical result, according to described voice signal
Level determine described speech loudness gain.
Volume control method the most according to claim 1, it is characterised in that described according to institute's predicate
The Gain generating of sonority gain and described largest tolerable adjusts gain, particularly as follows:
If the absolute value of described speech loudness gain is more than the second predetermined threshold value, then application below equation generates
Described adjustment gain G ain:
Gain=LGain × (1.0-(LGain+NGain)/(LGain × 2))；
Wherein, LGain is described speech loudness gain, and NGain is the gain of described largest tolerable.
Volume control method the most according to claim 1, it is characterised in that described according to described tune
Whole gain carries out Gain tuning to described acoustical signal, particularly as follows:
Determine adjustment duration according to described acoustical signal, determine according to described adjustment gain and described adjustment duration
Adjust step-length, according to described adjustment gain and described adjustment step-length, described acoustical signal is carried out Gain tuning.
Volume control method the most according to claim 1, it is characterised in that if described acoustical signal
Being at least two-way, the most described Gain generating according to described speech loudness gain and described largest tolerable is adjusted
After whole increasing, described according to described adjustment gain, described acoustical signal is carried out Gain tuning before, described
Method also includes:
Calculate described speech loudness gain and the difference of described adjustment gain on each road, determine in each road
Big difference, described adjustment gain G ain on each road is adjusted by application below equation:
Gain=2 × Gain LDiffMax LGain；
Wherein, LDiffMax is the difference of described maximum, and LGain is described speech loudness gain.
8. a loudness controls device, it is characterised in that including:
Detector unit, for carrying out speech detection respectively at least one road acoustical signal；
First processing unit, is connected with described detector unit, for for each road acoustical signal, according to
The voice signal detected, determines the speech loudness gain of described voice signal, quiet according to detect
Signal, determines the gain of the largest tolerable of described mute signal；
Second processing unit, is connected with described first processing unit, for according to described speech loudness gain
Gain is adjusted with the Gain generating of described largest tolerable；
First adjustment unit, is connected with described second processing unit, is used for according to described adjustment gain institute
State acoustical signal and carry out Gain tuning；
Wherein, described detector unit includes:
First processes subelement, for for described each road acoustical signal, calculates described acoustical signal
Root-mean-square；
Second processes subelement, processes subelement with described first and is connected, for dividing according to described root-mean-square
Sheng Cheng signal envelope and noise envelope；
Judgment sub-unit, processes subelement with described second and is connected, be used for calculating described signal envelope and institute
State the ratio of noise envelope, if described ratio is more than the first predetermined threshold value, then described voice signal detected,
Otherwise, described mute signal is detected.
Loudness the most according to claim 8 controls device, it is characterised in that: described first processes list
Unit is specifically for the level according to mute signal described in described root mean square calculation, according to described mute signal
Level determines the gain of described largest tolerable.
Loudness the most according to claim 9 controls device, it is characterised in that: described first processes
Unit is specifically for the level Noise_Level of the application below equation described mute signal of calculating:
Noise_Level=0.99 × Noise_Level+0.01 × Ecur；
Wherein, Ecur is described root-mean-square.
11. loudness according to claim 8 control device, it is characterised in that: described first processes
Unit is specifically for carrying out loudness Filtering Processing to described voice signal, to the voice after loudness Filtering Processing
Signal carries out signal level statistics, determines the level of described voice signal according to statistical result, according to described
The level of voice signal determines described speech loudness gain.
12. loudness according to claim 8 control device, it is characterised in that: described second processes
If unit is more than the second predetermined threshold value, then below application specifically for the absolute value of described speech loudness gain
Formula described adjustment gain G ain of generation:
Gain=LGain × (1.0-(LGain+NGain)/(LGain × 2))；
Wherein, LGain is described speech loudness gain, and NGain is the gain of described largest tolerable.
13. loudness according to claim 8 control device, it is characterised in that: described first adjusts
Unit is specifically for determining adjustment duration according to described acoustical signal, according to described adjustment gain and described tune
Whole duration determines adjustment step-length, enters described acoustical signal according to described adjustment gain and described adjustment step-length
Row Gain tuning.
14. loudness according to claim 8 control device, it is characterised in that if described sound letter
Number being at least two-way, described loudness controls device and also includes:
Second adjustment unit, is connected with described first adjustment unit, for calculating the described voice on each road
Loudness gain and the difference of described adjustment gain, determine difference maximum in each road, applies below equation pair
Described adjustment gain G ain on each road is adjusted:
Gain=2 × Gain LDiffMax LGain；
Wherein, LDiffMax is the difference of described maximum, and LGain is described speech loudness gain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210460201.0A CN103812462B (en) | 2012-11-15 | 2012-11-15 | Volume control method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210460201.0A CN103812462B (en) | 2012-11-15 | 2012-11-15 | Volume control method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103812462A CN103812462A (en) | 2014-05-21 |
| CN103812462B true CN103812462B (en) | 2016-12-07 |
Family
ID=50708755
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210460201.0A Active CN103812462B (en) | 2012-11-15 | 2012-11-15 | Volume control method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103812462B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105450193A (en) * | 2014-08-28 | 2016-03-30 | 深圳Tcl新技术有限公司 | Volume adjusting method and volume adjusting device |
| CN106992003A (en) * | 2017-03-24 | 2017-07-28 | 深圳北斗卫星信息科技有限公司 | Voice signal auto gain control method |
| CN108882115B (en) * | 2017-05-12 | 2020-08-25 | 华为技术有限公司 | Loudness adjustment method and device and terminal |
| CN107994879B (en) * | 2017-12-04 | 2022-07-08 | 北京小米移动软件有限公司 | Loudness control method and device |
| CN108806710B (en) * | 2018-06-15 | 2020-07-24 | 会听声学科技（北京）有限公司 | Voice enhancement gain adjustment method, system and earphone |
| CN116168719A (en) * | 2022-12-26 | 2023-05-26 | 杭州爱听科技有限公司 | Sound gain adjusting method and system based on context analysis |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783656A (en) * | 2010-03-17 | 2010-07-21 | 北京爱德发科技有限公司 | Loudness control method, module and device of stereo system |
| CN102436821A (en) * | 2011-12-02 | 2012-05-02 | 海能达通信股份有限公司 | Method for adaptively adjusting sound effect and equipment thereof |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW200620817A (en) * | 2004-12-10 | 2006-06-16 | Quanta Comp Inc | Audio output method and device for portable communication apparatus |
| US20090103751A1 (en) * | 2007-10-22 | 2009-04-23 | Stephen Gordon Lenk | Sound volume leveler for speed sensitive volume |
-
2012
- 2012-11-15 CN CN201210460201.0A patent/CN103812462B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101783656A (en) * | 2010-03-17 | 2010-07-21 | 北京爱德发科技有限公司 | Loudness control method, module and device of stereo system |
| CN102436821A (en) * | 2011-12-02 | 2012-05-02 | 海能达通信股份有限公司 | Method for adaptively adjusting sound effect and equipment thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103812462A (en) | 2014-05-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103812462B (en) | Volume control method and device | |
| CN102075842B (en) | Tinnitus hearing aid | |
| CN105376668B (en) | A kind of earphone noise-reduction method and device | |
| US10271138B2 (en) | Method and device for compensating frequency response of audio signal | |
| CN102860046A (en) | A hearing aid and a method for alleviating tinnitus | |
| CN102610229B (en) | Method, apparatus and device for audio dynamic range compression | |
| EP3264799A1 (en) | A method and a hearing device for improved separability of target sounds | |
| CN108573709B (en) | Automatic gain control method and device | |
| CN103473005A (en) | Method and device for performing sound effect control on played audio frequency | |
| CN102868962A (en) | Wireless binaural compressor | |
| CN105596008A (en) | Method of fitting a hearing device to a user, a fitting system for a hearing device and a hearing device | |
| CN107454537B (en) | Hearing device comprising a filter bank and an onset detector | |
| CN108882115A (en) | loudness adjusting method, device and terminal | |
| US11605392B2 (en) | Automatic gain control based on machine learning level estimation of the desired signal | |
| CN106409309A (en) | Tone quality enhancement method and microphone | |
| CN109803218B (en) | Automatic calibration method and device for loudspeaker sound field balance | |
| CN107948869A (en) | Audio-frequency processing method, device, sound system and storage medium | |
| WO2015085946A1 (en) | Voice signal processing method, apparatus and server | |
| CN109951784A (en) | Hearing devices and method with intelligently guiding | |
| CN113747336B (en) | Method for tuning and adapting sound field in different spaces based on audio processor | |
| CN113259801B (en) | Horn noise reduction method and related device of intelligent earphone | |
| CN112333602B (en) | Signal processing method, signal processing apparatus, computer-readable storage medium, and indoor playback system | |
| CN103248845A (en) | Voice-self adaptive television | |
| CN111063363A (en) | Voice acquisition method, audio equipment and device with storage function | |
| CN106796782A (en) | Information processor, information processing method and computer program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |