CN109215676B - Speech device with noise elimination and double-microphone speech system - Google Patents
Speech device with noise elimination and double-microphone speech system Download PDFInfo
- Publication number
- CN109215676B CN109215676B CN201710549873.1A CN201710549873A CN109215676B CN 109215676 B CN109215676 B CN 109215676B CN 201710549873 A CN201710549873 A CN 201710549873A CN 109215676 B CN109215676 B CN 109215676B
- Authority
- CN
- China
- Prior art keywords
- signal
- noise
- voice
- speech
- filter
- 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
- 230000008030 elimination Effects 0.000 title abstract description 5
- 238000003379 elimination reaction Methods 0.000 title abstract description 5
- 238000001914 filtration Methods 0.000 claims description 13
- 230000001755 vocal effect Effects 0.000 claims description 5
- 230000007423 decrease Effects 0.000 claims description 4
- 230000000717 retained effect Effects 0.000 abstract description 21
- 238000010586 diagram Methods 0.000 description 8
- 101000893549 Homo sapiens Growth/differentiation factor 15 Proteins 0.000 description 7
- 101000692878 Homo sapiens Regulator of MON1-CCZ1 complex Proteins 0.000 description 7
- 102100026436 Regulator of MON1-CCZ1 complex Human genes 0.000 description 7
- 102000008482 12E7 Antigen Human genes 0.000 description 4
- 108010020567 12E7 Antigen Proteins 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 238000000034 method Methods 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000009977 dual effect Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000005236 sound signal Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
- G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
- G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
- G10L21/0208—Noise filtering
- G10L21/0216—Noise filtering characterised by the method used for estimating noise
- G10L21/0232—Processing in the frequency domain
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04R—LOUDSPEAKERS, MICROPHONES, GRAMOPHONE PICK-UPS OR LIKE ACOUSTIC ELECTROMECHANICAL TRANSDUCERS; DEAF-AID SETS; PUBLIC ADDRESS SYSTEMS
- H04R3/00—Circuits for transducers, loudspeakers or microphones
Landscapes
- Engineering & Computer Science (AREA)
- Acoustics & Sound (AREA)
- Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Human Computer Interaction (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Computational Linguistics (AREA)
- Multimedia (AREA)
- Circuit For Audible Band Transducer (AREA)
- Noise Elimination (AREA)
- Telephone Function (AREA)
Abstract
The present disclosure provides a speech device with noise elimination and a dual-microphone speech system, wherein a speech receiver and a noise receiver simultaneously receive a first sound receiving source and a second sound receiving source to respectively obtain a first main signal with a human voice frequency band and a second main signal with a sudden noise frequency band. The speech filter filters the speech signal in the first host signal to preserve the bursty noise signal with a somewhat noisy speech signal (i.e., to preserve noise). The noise filter filters the burst noise signal in the second main signal to retain the speech signal with some burst noise signal (i.e., to retain speech). The post-filter generates a noise reduction gain based on the retained speech and the retained noise, and adjusts the retained speech based on the noise reduction gain to generate the speech signal. The speech device and the dual-microphone speech system can stably eliminate nearby noise and simultaneously keep clean external speech.
Description
Technical Field
The present disclosure relates to a speech apparatus and a dual-microphone speech system, and more particularly, to a speech apparatus and a dual-microphone speech system for eliminating a burst noise signal representing a keyboard tone or a key tone and retaining a speech signal representing a human voice.
Background
The prior art speech device uses a microphone to receive external speech (e.g. human voice) and nearby noise (e.g. environmental noise, key tone, keyboard tone, etc.), and eliminates the nearby noise by calculation, so as to generate clean speech. More specifically, when a microphone receives external voice and nearby noise to generate a mixed voice to a voice device, the voice device in the prior art uses Voice Activity Detection (VAD) and adaptive filter (adaptive filter) to remove the nearby noise and generate clean voice through a post-filter.
However, the voice device of the prior art only uses one sound signal, i.e. mixed sound (including external voice and nearby noise) to eliminate nearby noise, which easily causes the noise reduction effect to be unstable. Therefore, if the noise reduction effect can be improved, cleaner voice can be generated.
Disclosure of Invention
The invention aims to provide a voice device with noise elimination and a dual-microphone voice system, which utilize dual microphones to receive external voice and nearby noise (such as keyboard voice) so as to respectively generate two mixed sounds. The speech device and the dual-microphone speech system of the invention judge and process the two mixed sounds so as to stably eliminate the nearby noise and simultaneously keep clean external speech.
In one embodiment, the present disclosure provides a speech apparatus with noise cancellation for canceling a sudden noise signal and retaining a speech signal. The voice device with noise elimination comprises a voice detector, a noise detector, a voice filter, a noise filter and a post filter. The voice detector receives a first sound receiving source generated by a first microphone and a second sound receiving source generated by a second microphone. The voice detector acquires a human voice frequency band of the first sound receiving source as a first main signal, and generates a first result signal when the first main signal is judged to have the voice signal. The noise detector receives the first and second sound sources. The noise detector obtains a burst noise frequency band of the second sound receiving source as a second main signal, and generates a second result signal when judging that the second main signal has the burst noise signal. The voice filter is coupled to the voice detector and calculates a remaining noise according to the first result signal, the first main signal and the first reception source. The noise filter is coupled to the noise detector, and calculates a reserved voice according to the second result signal, the second main signal and the second radio source. The post-filter is coupled to the voice filter and the noise filter. The post-filter generates a noise reduction gain according to the retained speech and the retained noise, and generates a speech signal according to the noise reduction gain and the retained speech. If the post-filter judges that the first main signal has the voice signal, the post-filter maintains or increases the noise reduction gain. If the post-filter judges that the first main signal does not have the voice signal, the post-filter reduces the noise reduction gain.
Preferably, the first microphone and the second microphone have a predetermined distance therebetween.
Preferably, the voice detector comprises a band-pass filter and a voice judger. The band-pass filter acquires the voice frequency band of the first sound receiving source as a first main signal, and acquires the voice frequency band of the second sound receiving source as a first auxiliary signal. The voice judger is electrically connected with the band-pass filter and compares the energy of the first main signal with the energy of the first auxiliary signal. If the energy of the first main signal is higher than the energy of the first auxiliary signal to a predetermined value, the voice judger judges that the first main signal has a voice signal and generates a first result signal.
Preferably, the noise detector comprises a high pass filter and a noise determiner. The high-pass filter acquires the burst noise frequency band of the first sound receiving source as a second main signal, and acquires the burst noise frequency band of the second sound receiving source as a second auxiliary signal. The noise judger is electrically connected with the high-pass filter and compares the energy of the second main signal with the energy of the second auxiliary signal. If the energy of the second main signal is higher than the energy of the second auxiliary signal to a predetermined value, the noise determiner determines that the second main signal has a burst noise signal and generates a second result signal.
Preferably, the noise filter includes a second processor and a noise switch. The second processor receives a second sound source. The noise switch is electrically connected with the second processor and is switched on and off according to the second result signal. When the noise detector generates a second result signal, the noise switch is turned on, the second processor receives the second main signal, and a difference value between the second sound source and the second main signal is used as a reserved voice.
Preferably, the post-filter comprises a time-to-frequency domain converter, a computation component, a post-filter processor and a frequency-to-time domain converter. The time-frequency domain converter converts the reserved voice and the reserved noise from a time domain to a frequency domain to respectively generate a frequency domain voice and a frequency domain noise. The computing component is electrically connected with the time domain-frequency domain converter and computes a noise reduction gain between the frequency domain voice and the frequency domain noise. The post-filtering processor is electrically connected with the time domain-frequency domain converter, the calculating component and the voice detector, adjusts the noise reduction gain according to the first result signal or the second result signal, and adjusts the frequency domain voice according to the adjusted noise reduction gain to generate a voice adjusting signal. The frequency domain-time domain converter is electrically connected with the post-filtering processor and converts the voice adjusting signal from a frequency domain to a time domain so as to generate a voice signal.
Preferably, if the post-filtering processor determines that the first main signal has the voice signal according to the first result signal or the second result signal, the post-filtering processor maintains or increases the noise reduction gain. And if the post-filtering processor judges that the first main signal does not have the voice signal according to the first result signal or the second result signal, the post-filtering processor reduces the noise reduction gain.
Preferably, the sudden noise signal is a keyboard tone or a key tone.
In another embodiment, the present disclosure provides a dual-microphone speech system with noise cancellation for canceling a sudden noise signal representing a keyboard tone or a key tone and retaining a speech signal representing a human voice. The dual-microphone speech system with noise cancellation includes a first microphone, a second microphone, a speech detector, a noise detector, a speech filter, a noise filter, and a post-filter. The first microphone and the second microphone receive a voice signal generated by a voice source and a burst noise signal generated by a noise source to respectively generate a first sound receiving source and a second sound receiving source. The voice detector is coupled with the first microphone and the second microphone. The voice detector receives the first sound receiving source and the second sound receiving source, obtains a human voice frequency band of the first sound receiving source as a first main signal, and generates a first result signal when the first main signal is judged to have the voice signal. The noise detector is coupled to the first microphone and the second microphone. The noise detector receives the first sound receiving source and the second sound receiving source, obtains a burst noise frequency band of the second sound receiving source as a second main signal, and generates a second result signal when judging that the second main signal has the burst noise signal. The voice filter is coupled to the voice detector and calculates a remaining noise according to the first result signal, the first main signal and the first reception source. The noise filter is coupled to the noise detector, and calculates a reserved voice according to the second result signal, the second main signal and the second radio source. The post filter is coupled to the voice filter and the noise filter, and generates a noise reduction gain according to the reserved voice and the reserved noise, and generates a voice signal according to the noise reduction gain and the reserved voice. If the post-filter judges that the first main signal has the voice signal, the post-filter maintains or increases the noise reduction gain. If the post-filter judges that the first main signal does not have the voice signal, the post-filter reduces the noise reduction gain.
To further clarify the features and technical content of the present disclosure, reference is made to the following detailed description and accompanying drawings of the present disclosure, which are provided for the purpose of illustration only and are not intended to limit the scope of the present disclosure.
Drawings
Fig. 1 is a schematic diagram of a user-operated dual-microphone speech system provided by an embodiment of the present disclosure.
Fig. 2 is a schematic diagram of a speech device provided in an embodiment of the present disclosure.
Fig. 3A is a schematic diagram of a speech detector provided by an embodiment of the present disclosure.
Fig. 3B is a schematic diagram of a noise detector provided in an embodiment of the disclosure.
Fig. 4A is a schematic diagram of a speech filter provided by an embodiment of the present disclosure.
Fig. 4B is a schematic diagram of a noise filter provided by an embodiment of the disclosure.
Fig. 5 is a schematic diagram of a post-filter provided by an embodiment of the disclosure.
Detailed Description
Hereinafter, the present disclosure will be described in detail by way of illustration of the drawings and various exemplary embodiments of the present disclosure. The disclosed concepts may, however, be embodied in many different forms and should not be construed as limited to the exemplary embodiments set forth herein. Moreover, in the drawings, like reference numerals may be used to designate similar components.
In the speech device with noise cancellation and the dual-microphone speech system provided in the embodiments of the present disclosure, the speech receiver and the noise receiver simultaneously receive a first sound receiving source (including external speech (e.g., voice) and nearby noise (e.g., keyboard tone and key tone)) and a second sound receiving source (including external speech (e.g., voice) and nearby noise (e.g., keyboard tone and key tone)) to respectively obtain a first main signal with a voice frequency band and a second main signal with a burst noise frequency band. The voice filter filters the external voice according to the first result signal, the first main signal and the first receiving source to keep noise (with little voice). The noise filter will filter the nearby noise according to the second result signal, the second main signal and the second radio source to retain the voice (with some noise). The filter generates a noise reduction gain according to the reserved voice and the reserved noise, and adjusts the reserved voice according to the noise reduction gain to generate a clean voice signal. Therefore, the voice device and the dual-microphone voice system of the invention judge and process the first sound receiving source and the second sound receiving source so as to stably eliminate nearby noise and simultaneously keep clean external voice. The speech device with noise cancellation and the two-microphone speech system provided by the embodiments of the present disclosure will be further described below.
First, please refer to fig. 1, which respectively shows a schematic diagram of a user-operated dual-microphone speech system according to an embodiment of the present disclosure. As shown in fig. 1, the two-microphone speech system includes a speech device 100 and a two-microphone 110. Two microphones 110 are coupled to speech device 100. The dual microphone 110 includes a first microphone MIC1 and a second microphone MIC 2. In this embodiment, the audio apparatus 100 may be a computer host, a smart phone or other electronic apparatuses, which is not limited in this disclosure. The audio device 100 is coupled to the keyboard 60 and the screen 70, and is connected to the remote electronic device 50 through a wireless network. The first microphone MIC1 and the second microphone MIC2 have a predetermined distance therebetween. The user (a speech source in this embodiment) is closer to the first microphone MIC1 and further from the second microphone MIC 2. The keyboard 60 (which is a source of noise in this embodiment) is closer to the second microphone MIC2 and further from the first microphone MIC 1.
When a user speaks and types (i.e., taps the keyboard 60), the first microphone MIC1 and the second microphone MIC2 receive a voice signal generated by the user (i.e., the voice source in this embodiment) and an abrupt noise signal generated by the keyboard 60 (i.e., the noise source in this embodiment), and respectively generate a first sound receiving source M1 and a second sound receiving source M2 to the voice apparatus 100. Due to the arrangement relationship among the user, the keyboard 60, the first microphone MIC1 and the second microphone MIC2, when the user speaks while typing, the voice signal of the first microphone M1 is larger than the sudden noise signal, and the voice signal VOC of the second microphone M2 is smaller than the sudden noise signal. When the user only types, the voice signal of the first sound source M1 will be smaller than the sudden noise signal, and the voice signal VOC of the second sound source M2 will be smaller than the sudden noise signal.
The voice device 100 is used to eliminate the sudden noise signal generated by the noise source and to retain the voice signal VOC generated by the voice source. The speech device 100 displays the typed content of the user on the screen 70 and transmits the content to the remote electronic device 50 for viewing by the remote user. The speech device 100 also transmits the content of the user's speech (i.e. the speech signal VOC) to the remote electronic device for the remote user to listen to.
More specifically, as shown in fig. 2, the speech apparatus 100 has a speech detector 120, a noise detector 130, a speech filter 140, a noise filter 150 and a post-filter 160. The voice detector 120 receives a first microphone M1 from the first microphone MIC1 and a second microphone M2 from the second microphone MIC 2. The voice detector 120 obtains a human voice frequency band of the first sound source M1 as a first main signal Sv1, and generates a first result signal R1 when determining that the first main signal Sv1 has the voice signal VOC.
Referring to fig. 3A, in the present embodiment, the voice detector 120 includes a band-pass filter 122 and a voice determiner 124. The band-pass filter 122 obtains the vocal band of the first sound source M1 as the first main signal Sv1, and obtains the vocal band of the second sound source M2 as the first auxiliary signal Sv 2. The human voice frequency band of the embodiment is a 300Hz-2kHz frequency band. Therefore, the first main signal Sv1 is the signal of the first sound source M1 in the 300Hz-2kHz frequency band. The first auxiliary signal Sv2 is the signal of the second sound source M2 in the frequency band of 300Hz-2 kHz. Of course, the human frequency band may be set to other suitable frequency bands according to actual conditions, which is not limited in the present invention.
The voice determiner 124 is electrically connected to the band pass filter 122, and compares the energy of the first main signal Sv1 with the energy of the first auxiliary signal Sv 2. If the energy of the first main signal Sv1 is higher than the energy of the first auxiliary signal Sv2 by a predetermined value, the voice determinator 124 determines that the voice signal VOC is present in the first main signal Sv1 and generates a first result signal R1 correspondingly. For example, the energy of the first main signal Sv1 is divided by the energy of the first auxiliary signal Sv2, and the calculation result is greater than the value 2, which indicates that the energy of the first main signal Sv1 is higher than the energy of the first auxiliary signal Sv2 to a predetermined value. At this time, the voice determiner 124 will generate the first result signal R1 with a high level. On the contrary, if the energy of the first main signal Sv1 is lower than the energy of the first auxiliary signal Sv2 to a predetermined value, the voice determinator 124 determines that the first main signal Sv1 does not have the voice signal VOC and does not generate the first result signal R1, i.e., the low-level first result signal R1.
Similarly, referring to fig. 2 and fig. 3B, the noise detector 130 receives the first microphone M1 from the first microphone MIC1 and the second microphone M2 from the second microphone MIC 2. The noise detector 130 obtains a burst noise band of the second sound source M2 as a second main signal Sn1, and generates a second result signal R2 when determining that the second main signal Sn1 has the burst noise signal.
In the present embodiment, the noise detector 130 includes a high pass filter 132 and a noise determiner 134. The high pass filter 132 obtains the noise burst band of the second sound source M2 as the second main signal Sn1, and obtains the noise burst band of the first sound source M1 as a second auxiliary signal Sn 2. The burst noise frequency band of the present embodiment is a frequency band above 4 kHz. Therefore, the second main signal Sn1 is a signal of the second sound source M2 in the frequency band of 4kHz or more. The second auxiliary signal Sn2 is a signal of the first sound source M1 in the frequency band above 4 kHz. Of course, the burst noise frequency band may be set to other suitable frequency bands according to actual conditions, which is not limited in the present invention.
The noise determiner 134 is electrically connected to the high pass filter 132, and compares the energy of the second main signal Sn1 with the energy of the second auxiliary signal Sn 2. If the energy of the second main signal Sn1 is higher than the energy of the second auxiliary signal Sn2 by a predetermined value, the noise determiner 134 determines that the second main signal Sn1 has a sudden noise signal and correspondingly generates a second result signal R2. For example, the energy of the second main signal Sn1 is divided by the energy of the second auxiliary signal Sn2, and the calculation result is greater than 2, which indicates that the energy of the second main signal Sn1 is higher than the energy of the second auxiliary signal Sn2 to a predetermined value. At this time, the noise determiner 134 will generate the second result signal R2 with a high level. On the contrary, if the energy of the second main signal Sn1 is lower than the energy of the second auxiliary signal Sn2 to a predetermined value, the noise determiner 134 determines that the second main signal Sn1 does not have the pop noise signal and does not generate the second result signal R2, i.e., the low level second result signal R2.
Next, referring to fig. 2 and fig. 4A, the voice filter 140 is coupled to the voice detector 120, and calculates a noise retention Ref according to the first result signal R1, the first main signal Sv1 and the first sound source M1. Further, as shown in fig. 4A, the voice filter 140 includes a first processor 144 and a voice switch 142. The first processor 144 receives the first sound source M1. The voice switch 142 is electrically connected to the first processor 144 and is turned on or off according to the first result signal R1.
When the voice detector 120 generates the first result signal R1 (in this embodiment, the first result signal R1 with a high level), the voice switch 142 is turned on. At this time, the first processor 144 receives the first main signal Sv1 and takes a difference between the first sound source M1 and the first main signal Sv1 as a remaining noise Ref. At this point, the remaining noise Ref has a complete burst noise signal and a small portion of the speech signal VOC. On the contrary, when the voice detector 120 does not generate the first result signal R1 (in this embodiment, the first result signal R1 of low level), the voice switch 142 is turned off. At this time, the first processor 144 does not calculate the remaining noise Ref (the remaining noise Ref at a low level in the present embodiment).
Similarly, referring to fig. 2 and fig. 4B, the noise filter 150 is coupled to the noise detector 130, and calculates a remaining voice Tar according to the second result signal R2, the second main signal Sn1 and the second sound source M2. More specifically, as shown in fig. 4B, the noise filter 150 includes a second processor 154 and a noise switch 152. The second processor 154 receives a second sound source M2. The noise switch 152 is electrically connected to the second processor 154 and is turned on and off according to the second result signal R2.
When the noise detector 130 generates the second result signal R2, (in this embodiment, the second result signal R2 with a high level), the noise switch 152 is turned on. The second processor 154 receives the second main signal Sn1 and takes a difference between the second sound source M2 and the second main signal Sn1 as a reserved speech Tar. At this time, the remaining voice Tar has the complete voice signal VOC and a small portion of the sudden noise signal. On the contrary, when the noise detector 130 does not generate the second result signal R2 (in this embodiment, the second result signal R2 with a low level), the noise switch 152 is turned off. At this time, the second processor 154 does not calculate the retained speech Tar (low-level retained speech Tar in the present embodiment).
Referring to fig. 2 and 5, the post-filter 160 is coupled to the voice filter 140 and the noise filter 150, and generates a noise reduction gain according to the retained voice Tar and the retained noise Ref, and generates the voice signal VOC according to the noise reduction gain and the retained voice Tar. If the post-filter 160 determines that the first main signal Sv1 has the voice signal VOC, the post-filter 160 maintains or increases the noise reduction gain. If the post-filter 160 determines that the first main signal Sv1 does not have the voice signal VOC, the post-filter 160 decreases the noise reduction gain.
Further, as shown in fig. 5, the post-filter 160 includes a time-to-frequency domain converter 162, a calculating component 164, a post-filter processor 166 and a frequency-to-time domain converter 168. The time-to-frequency converter 162 converts the retained speech Tar and the retained noise Ref from a time domain to a frequency domain to generate a frequency domain speech Tf and a frequency domain noise Rf, respectively. The embodiments of time-domain to frequency-domain conversion are well known in the art and therefore will not be described herein.
The computation component 164 is electrically connected to the time-frequency domain converter 162. The calculation component 164 receives the retained speech Tar and the retained noise Ref and calculates a noise reduction gain Gn between the frequency domain speech Tar and the frequency domain noise Ref. In the present embodiment, the noise reduction gain Gn is a signal-to-noise ratio (SNR). Therefore, the noise reduction gain Gn is a ratio of the retained speech Tar to the retained noise Ref. The noise reduction gain Gn may also be calculated according to actual conditions, which is not limited in the present invention.
The post-filter processor 166 electrically connects the time-to-frequency domain converter 162, the computation component 164, and the speech detector 120. The post-filter processor 166 adjusts the noise reduction gain Gn according to the first result signal R1, and adjusts the frequency domain speech Tf according to the adjusted noise reduction gain Gn to generate a speech adjustment signal Tf'. Further, when the voice detector 120 generates the first result signal R1 (in this embodiment, the first result signal R1 with a high level), it indicates that the user is speaking. At this time, the post-filter processor 166 will determine that the voice signal VOC is present in the first main signal Sv1 according to the first result signal R1. The post-filter processor 166 maintains or increases the noise reduction gain Gn and correspondingly adjusts the frequency-domain speech Tf to generate the speech adjustment signal Tf'. On the contrary, when the voice detector 120 does not generate the first result signal R1 (the first result signal R1 with a low level in this embodiment), it indicates that the user is not speaking. At this time, the post-filter processor 166 will determine from the first result signal R1 that the first main signal Sv1 does not have the voice signal VOC. The post-filter processor 166 will adjust the noise reduction gain Gn low and correspondingly adjust the frequency domain speech Tf to generate the speech adjustment signal Tf'.
In other embodiments, the post-filtering processor 166 can also adjust the noise reduction gain Gn according to the second result signal R2, and adjust the frequency domain speech Tf according to the adjusted noise reduction gain Gn to generate the speech adjusting signal Tf'. For example, if the noise detector 130 generates the second result signal R2 (in this embodiment, the second result signal R2 with a high level), it indicates that the user is interfered by the pop noise signal. At this time, the post-filtering processor 166 will lower the noise reduction gain Gn according to the second result signal R2 and correspondingly adjust the frequency-domain speech Tf to generate the speech adjusting signal Tf'. On the contrary, if the noise detector 130 does not generate the second result signal R2 (in this embodiment, the second result signal R2 with a low level), it indicates that the user is not interfered by the pop noise signal. At this time, the post-filtering processor 166 maintains or increases the noise reduction gain Gn according to the second result signal R2 and correspondingly adjusts the frequency-domain speech Tf to generate the speech adjustment signal Tf'.
After the speech adjustment signal Tf 'is generated, a frequency-to-time domain converter 168, which is electrically connected to the post-filter processor 166, converts the speech adjustment signal Tf' from the frequency domain to the time domain to generate the speech signal VOC therefrom. The embodiments related to converting frequency domain to time domain are well known in the art, and therefore will not be described herein.
In summary, in the speech apparatus with noise cancellation and the dual-microphone speech system provided in the embodiments of the invention, the speech receiver 120 and the noise receiver 130 simultaneously receive the first receiving source M1 and the second receiving source M2 to respectively obtain the first main signal Sv1 with the vocal band and the second main signal Sn1 with the burst noise band. The voice filter 140 filters the voice signal according to the first result signal R1, the first main signal Sv1 and the first sound source M1 to retain the noise burst signal with a slight voice signal (i.e., the retained noise Ref). The noise filter 150 will filter the noise burst signal according to the second result signal R2, the second main signal Sn1 and the second sound source M2 to generate the speech signal with some noise burst signal (i.e. the remaining speech Tar). Then, the filter 160 will generate a noise reduction gain Gn according to the retained speech Tar and the retained noise Ref, and adjust the retained speech Tar according to the noise reduction gain Gn to generate the speech signal VOC. Accordingly, the speech apparatus and the dual-microphone speech system of the present invention determine and process the first sound source M1 and the second sound source M2 to stably eliminate the burst noise signal and simultaneously maintain a clean speech signal.
The above description is only for the best mode of the present disclosure, but the present disclosure is not limited thereto, and any changes or modifications that can be easily made by those skilled in the art within the field of the present disclosure can be covered by the claims of the present disclosure.
Claims (10)
1. A speech apparatus with noise cancellation for canceling a sudden noise signal and preserving a speech signal, comprising:
a voice detector, receiving a first receiving source generated by a first microphone and a second receiving source generated by a second microphone, obtaining a human voice frequency band of the first receiving source as a first main signal and obtaining the human voice frequency band of the second receiving source as a first auxiliary signal, and comparing the energy of the first main signal with the energy of the first auxiliary signal, and if the energy of the first main signal is higher than the energy of the first auxiliary signal to a predetermined value, determining that the voice signal exists in the first main signal to generate a first result signal, wherein the first microphone is close to a voice source generating the voice signal, and the second microphone is close to a noise source generating the burst noise signal;
a noise detector for receiving the first and second sound receiving sources, obtaining a burst noise frequency band of the second sound receiving source as a second main signal and obtaining the burst noise frequency band of the first sound receiving source as a second auxiliary signal, and comparing the energy of the second main signal with the energy of the second auxiliary signal, and if the energy of the second main signal is higher than the energy of the second auxiliary signal to another predetermined value, determining that the second main signal has the burst noise signal to generate a second result signal;
a voice filter coupled to the voice detector and calculating a remaining noise according to the first result signal, the first main signal and the first receiving source;
a noise filter coupled to the noise detector and calculating a reserved voice according to the second result signal, the second main signal and the second radio source; and
a post filter coupled to the voice filter and the noise filter, generating a noise reduction gain according to the reserved voice and the reserved noise, and generating the voice signal according to the noise reduction gain and the reserved voice;
if the post-filter determines that the first main signal has the voice signal, the post-filter maintains or increases the noise reduction gain, and if the post-filter determines that the first main signal does not have the voice signal, the post-filter decreases the noise reduction gain.
2. The audio device according to claim 1, wherein the first microphone is spaced apart from the second microphone by a predetermined distance.
3. The speech device with noise cancellation of claim 1, wherein the speech detector comprises:
a band-pass filter for obtaining the vocal band of the first sound receiving source as the first main signal and obtaining the vocal band of the second sound receiving source as the first auxiliary signal; and
and the voice judger is electrically connected with the band-pass filter, compares the energy of the first main signal with the energy of the first auxiliary signal, judges that the voice signal exists in the first main signal and generates the first result signal if the energy of the first main signal is higher than the energy of the first auxiliary signal to the preset value.
4. The speech device with noise cancellation of claim 1, wherein the noise detector comprises:
a high-pass filter for obtaining the burst noise frequency band of the second sound-receiving source as the second main signal and obtaining the burst noise frequency band of the first sound-receiving source as the second auxiliary signal; and
and the noise judger is electrically connected with the high-pass filter, compares the energy of the second main signal with the energy of the second auxiliary signal, judges that the second main signal has the burst noise signal and generates a second result signal if the energy of the second main signal is higher than the energy of the second auxiliary signal to the other preset value.
5. The speech device with noise cancellation of claim 1, wherein the speech filter comprises:
a first processor for receiving the first sound source; and
a voice switch electrically connected to the first processor and conducting and stopping according to the first result signal;
when the voice detector generates the first result signal, the voice switch is turned on, the first processor receives the first main signal, and a difference value between the first sound receiving source and the first main signal is used as the reserved noise.
6. The speech device with noise cancellation of claim 1, wherein the noise filter comprises:
a second processor for receiving the second sound source; and
a noise switch electrically connected to the second processor and turned on and off according to the second result signal;
when the noise detector generates the second result signal, the noise switch is turned on, the second processor receives the second main signal, and a difference value between the second sound receiving source and the second main signal is used as the reserved voice.
7. The speech device with noise cancellation of claim 1, wherein the post-filter comprises:
a time-frequency domain converter, converting the reserved voice and the reserved noise from a time domain to a frequency domain to respectively generate a frequency domain voice and a frequency domain noise;
a calculating component, electrically connected to the time-frequency domain converter, for calculating a noise reduction gain between the frequency domain speech and the frequency domain noise;
a post-filter processor electrically connected to the time-domain-to-frequency-domain converter, the computing component and the voice detector, for adjusting the noise reduction gain according to the first result signal or the second result signal, and adjusting the frequency-domain voice according to the adjusted noise reduction gain to generate a voice adjustment signal; and
a frequency-time converter electrically connected to the post-filtering processor for converting the voice adjusting signal from a frequency domain to a time domain to generate the voice signal.
8. The speech device with noise cancellation as claimed in claim 7, wherein the post-filtering processor maintains or increases the noise reduction gain if the post-filtering processor determines that the speech signal exists in the first main signal according to the first result signal or determines that the burst noise signal does not exist in the second main signal according to the second result signal, and decreases the noise reduction gain if the post-filtering processor determines that the speech signal does not exist in the first main signal according to the first result signal or determines that the burst noise signal exists in the second main signal according to the second result signal.
9. The speech device with noise cancellation of claim 1, wherein the sudden noise signal is a keyboard tone or a touch tone.
10. A dual-microphone speech system with noise cancellation for canceling a sudden noise signal representing a keyboard tone or a key tone and retaining a speech signal representing a human voice, comprising:
a first microphone and a second microphone for receiving the voice signal generated by a voice source and the burst noise signal generated by a noise source to generate a first sound receiving source and a second sound receiving source respectively;
a voice detector, coupled to the first microphone and the second microphone, for receiving the first sound receiving source and the second sound receiving source, obtaining a human voice frequency band of the first sound receiving source as a first main signal and obtaining the human voice frequency band of the second sound receiving source as a first auxiliary signal, and comparing energy of the first main signal with energy of the first auxiliary signal, and if the energy of the first main signal is higher than the energy of the first auxiliary signal to a predetermined value, determining that the voice signal exists in the first main signal to generate a first result signal;
a noise detector, coupled to the first microphone and the second microphone, for receiving the first sound receiving source and the second sound receiving source, obtaining a burst noise frequency band of the second sound receiving source as a second main signal and obtaining the burst noise frequency band of the first sound receiving source as a second auxiliary signal, and comparing energy of the second main signal with energy of the second auxiliary signal, and if the energy of the second main signal is higher than the energy of the second auxiliary signal to another predetermined value, determining that the second main signal has the burst noise signal and generating a second result signal;
a voice filter coupled to the voice detector and calculating a remaining noise according to the first result signal, the first main signal and the first receiving source;
a noise filter coupled to the noise detector and calculating a reserved voice according to the second result signal, the second main signal and the second radio source; and
a post filter coupled to the voice filter and the noise filter, generating a noise reduction gain according to the reserved voice and the reserved noise, and generating the voice signal according to the noise reduction gain and the reserved voice;
if the post-filter determines that the first main signal has the voice signal, the post-filter maintains or increases the noise reduction gain, and if the post-filter determines that the first main signal does not have the voice signal, the post-filter decreases the noise reduction gain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710549873.1A CN109215676B (en) | 2017-07-07 | 2017-07-07 | Speech device with noise elimination and double-microphone speech system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710549873.1A CN109215676B (en) | 2017-07-07 | 2017-07-07 | Speech device with noise elimination and double-microphone speech system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109215676A CN109215676A (en) | 2019-01-15 |
| CN109215676B true CN109215676B (en) | 2021-05-18 |
Family
ID=64991067
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710549873.1A Active CN109215676B (en) | 2017-07-07 | 2017-07-07 | Speech device with noise elimination and double-microphone speech system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109215676B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109741762B (en) * | 2019-02-15 | 2020-12-22 | 嘉楠明芯(北京)科技有限公司 | Voice activity detection method and device and computer readable storage medium |
| CN112153505A (en) * | 2019-06-28 | 2020-12-29 | 中强光电股份有限公司 | Noise reduction system and noise reduction method |
| CN111370033B (en) * | 2020-03-13 | 2023-09-22 | 北京字节跳动网络技术有限公司 | Keyboard sound processing method and device, terminal equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5568559A (en) * | 1993-12-17 | 1996-10-22 | Canon Kabushiki Kaisha | Sound processing apparatus |
| CN1670823A (en) * | 2004-03-17 | 2005-09-21 | 哈曼贝克自动系统股份有限公司 | Method for detecting and reducing noise from a microphone array |
| CN1832636A (en) * | 2005-03-01 | 2006-09-13 | 奥迪康有限公司 | System and method for determining directionality of sound detected by a hearing aid |
| JP4348000B2 (en) * | 2000-08-31 | 2009-10-21 | 三洋化成工業株式会社 | Base oil for lubricating oil |
| CN203882609U (en) * | 2014-05-08 | 2014-10-15 | 钰太芯微电子科技（上海）有限公司 | Awakening apparatus based on voice activation detection |
| CN105814914A (en) * | 2013-12-12 | 2016-07-27 | 株式会社索思未来 | Audio playback device and game device |
| CN106303838A (en) * | 2015-06-25 | 2017-01-04 | 宏达国际电子股份有限公司 | Sound processing apparatus and method |
| JP6316707B2 (en) * | 2014-08-26 | 2018-04-25 | 花王株式会社 | Pants-type absorbent article |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH06101666B2 (en) * | 1986-07-08 | 1994-12-12 | 日本電気株式会社 | Adaptive noise eliminator |
| JPS63292814A (en) * | 1987-05-26 | 1988-11-30 | Nec Corp | Adaptive type noise eliminator |
| JP2942034B2 (en) * | 1991-01-07 | 1999-08-30 | キヤノン株式会社 | Audio processing device |
| KR100911610B1 (en) * | 2007-07-20 | 2009-08-07 | 에스케이 텔레콤주식회사 | Voice Process Apparatus and Method for Receipt Voice Recognition Ratio Improvement |
-
2017
- 2017-07-07 CN CN201710549873.1A patent/CN109215676B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5568559A (en) * | 1993-12-17 | 1996-10-22 | Canon Kabushiki Kaisha | Sound processing apparatus |
| JP4348000B2 (en) * | 2000-08-31 | 2009-10-21 | 三洋化成工業株式会社 | Base oil for lubricating oil |
| CN1670823A (en) * | 2004-03-17 | 2005-09-21 | 哈曼贝克自动系统股份有限公司 | Method for detecting and reducing noise from a microphone array |
| CN1832636A (en) * | 2005-03-01 | 2006-09-13 | 奥迪康有限公司 | System and method for determining directionality of sound detected by a hearing aid |
| CN105814914A (en) * | 2013-12-12 | 2016-07-27 | 株式会社索思未来 | Audio playback device and game device |
| CN203882609U (en) * | 2014-05-08 | 2014-10-15 | 钰太芯微电子科技（上海）有限公司 | Awakening apparatus based on voice activation detection |
| JP6316707B2 (en) * | 2014-08-26 | 2018-04-25 | 花王株式会社 | Pants-type absorbent article |
| CN106303838A (en) * | 2015-06-25 | 2017-01-04 | 宏达国际电子股份有限公司 | Sound processing apparatus and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109215676A (en) | 2019-01-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10339952B2 (en) | Apparatuses and systems for acoustic channel auto-balancing during multi-channel signal extraction | |
| US8194881B2 (en) | Detection and suppression of wind noise in microphone signals | |
| KR102205574B1 (en) | Systems and methods for detection and cancellation of narrow-band noise | |
| US8903721B1 (en) | Smart auto mute | |
| KR101275442B1 (en) | Systems, methods, apparatus, and computer-readable media for phase-based processing of multichannel signal | |
| US7680465B2 (en) | Sound enhancement for audio devices based on user-specific audio processing parameters | |
| US9711162B2 (en) | Method and apparatus for environmental noise compensation by determining a presence or an absence of an audio event | |
| US9530427B2 (en) | Speech processing | |
| US20130218559A1 (en) | Noise reduction apparatus, audio input apparatus, wireless communication apparatus, and noise reduction method | |
| CN109215676B (en) | Speech device with noise elimination and double-microphone speech system | |
| CN112424863A (en) | Voice perception audio system and method | |
| WO2007014136A9 (en) | Robust separation of speech signals in a noisy environment | |
| CN111316356A (en) | Adaptive null shaping for selective audio pickup | |
| EP3340657B1 (en) | A hearing device comprising a dynamic compressive amplification system and a method of operating a hearing device | |
| US20160035367A1 (en) | Speech dereverberation methods, devices and systems | |
| CA2798282A1 (en) | Wind suppression/replacement component for use with electronic systems | |
| JP2008197200A (en) | Automatic intelligibility adjusting device and automatic intelligibility adjusting method | |
| US10083707B1 (en) | Voice apparatus and dual-microphone voice system with noise cancellation | |
| US9532138B1 (en) | Systems and methods for suppressing audio noise in a communication system | |
| JP2023159381A (en) | Sound recognition audio system and method thereof | |
| US10616676B2 (en) | Dynamically adjustable sidetone generation | |
| JP5903921B2 (en) | Noise reduction device, voice input device, wireless communication device, noise reduction method, and noise reduction program | |
| WO2013078677A1 (en) | A method and device for adaptively adjusting sound effect | |
| EP3830823B1 (en) | Forced gap insertion for pervasive listening | |
| US11323804B2 (en) | Methods, systems and apparatus for improved feedback control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |