WO2015176372A1 - 智能终端通话降噪方法及智能终端 - Google Patents
智能终端通话降噪方法及智能终端 Download PDFInfo
- Publication number
- WO2015176372A1 WO2015176372A1 PCT/CN2014/082285 CN2014082285W WO2015176372A1 WO 2015176372 A1 WO2015176372 A1 WO 2015176372A1 CN 2014082285 W CN2014082285 W CN 2014082285W WO 2015176372 A1 WO2015176372 A1 WO 2015176372A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- terminal
- noise
- voice
- quiet state
- ambient noise
- 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.)
- Ceased
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M1/00—Substation equipment, e.g. for use by subscribers
- H04M1/02—Constructional features of telephone sets
- H04M1/19—Arrangements of transmitters, receivers, or complete sets to prevent eavesdropping, to attenuate local noise or to prevent undesired transmission; Mouthpieces or receivers specially adapted therefor
Definitions
- the present invention relates to the field of communications technologies, and in particular, to a method and a smart terminal for noise reduction of an intelligent terminal.
- Intelligent terminals in particular, smart terminals (eg, smart watches, smart glasses, etc.) are used as a new type of smart terminal that can be conveniently worn on a specific body part of a person.
- Existing wearable smart terminals generally have a wireless calling function, which is usually used in a special environment, for example, when people's hands cannot hold or operate a mobile terminal (for example, a mobile phone, a tablet, etc.)
- a mobile terminal for example, a mobile phone, a tablet, etc.
- the use of such wearable smart terminals to communicate directly with other terminal devices greatly expands the scene of people's voice calls.
- the limited structure of the wearable smart terminal is relatively small, and an array composed of multiple MICs (microphones) cannot be set for denoising. Therefore, the call of the smart terminal is generally susceptible to environmental noise, especially In a noisy environment, the quality of the call cannot be guaranteed.
- the above content is only used to assist in understanding the technical solutions of the present invention, and does not constitute an admission that the above is prior art.
- the embodiments of the present invention provide a method for reducing noise of an intelligent terminal and an intelligent terminal, which are intended to solve the problem that the existing smart terminal, especially for the wearable smart terminal, cannot be denoised due to its small size.
- a method for call noise reduction of an intelligent terminal including: when a voice communication starts, a first terminal collects a call voice and an environmental noise; and the first terminal according to the call voice or/and environment collected by the first terminal a noise, determining whether the first terminal is in a quiet state; if the first terminal is not in a quiet state, the first terminal performs noise reduction processing on the call voice according to the ambient noise sent by the second terminal; If you are in a quiet state, do not do noise reduction.
- the step of the first terminal determining whether it is in a quiet state according to the call voice or/and the ambient noise collected by the first terminal includes: And obtaining a signal to noise ratio according to the call voice and the ambient noise collected by the first terminal; if the signal to noise ratio is greater than or equal to a preset value, the first terminal is in a quiet state.
- the step of determining, by the first terminal, whether the device is in a quiet state according to the collected call voice or/and the ambient noise includes: obtaining an amplitude of the ambient noise according to the ambient noise collected by the first terminal; If the noise amplitude value is less than the preset noise signal amplitude threshold, the first terminal is in a quiet state.
- the step of performing noise reduction processing on the call voice by the first terminal according to the ambient noise sent by the second terminal includes: if the first terminal is not in the In a quiet state, a voice link with the second terminal is established, and a microphone of the second terminal is started to collect ambient noise. The first terminal performs noise reduction processing according to the ambient noise sent by the second terminal.
- the first terminal is not in a quiet state for performing noise reduction processing, or the step of the first terminal being in a quiet state without performing noise reduction processing further includes: the first terminal real-time monitoring whether the environment in which the terminal is located is In the quiet state, it is determined whether the call voice needs to be noise-reduced; if the first terminal is in an unquiet state, the call voice is subjected to noise reduction processing; if the first terminal is in a quiet state, the call voice is not noise-reduced deal with.
- the method further includes: the first terminal establishing a persistent voice link with the second terminal, and the microphone of the second terminal The collected environmental noise is continuously sent to the first terminal.
- an intelligent terminal including: a sound collection unit configured to collect a call voice and ambient noise at the beginning of voice communication; a quiet state determining unit, configured to determine whether it is in a quiet state according to a call voice or/and an environmental noise collected by the main microphone; and a voice noise reduction output unit configured to be in a state where the first terminal is not in a quiet state, according to the The ambient noise sent by the second terminal performs noise reduction processing on the call voice, and the voice output unit is set to be in a quiet state in the first terminal, and does not perform noise reduction processing on the call voice.
- the quiet state determining unit is configured to: obtain a signal to noise ratio according to the call voice and the ambient noise collected by the first terminal; if the signal to noise ratio is greater than or equal to a preset value, the first terminal is at Preferably, the quiet state determining unit is configured to: obtain an amplitude of the ambient noise according to the ambient noise collected by the first terminal; and if the ambient noise amplitude value is less than a preset noise signal amplitude threshold, The first terminal is in a quiet state.
- the voice noise reduction output unit includes: a link enabling module, configured to establish a voice link with the second terminal when the first terminal is not in a quiet state, and start the second terminal
- the microphone collects ambient noise
- the noise reduction output module is configured to perform noise reduction processing on the call voice according to the ambient noise sent by the second terminal.
- the smart terminal further includes: a real-time monitoring unit, configured to monitor in real time whether the environment in which the first terminal is located is in a quiet state, determine whether the call voice needs to be noise-reduced; if the first terminal is not quiet In the state, the call voice is subjected to noise reduction processing; if the first terminal is in a quiet state, the call voice is not subjected to noise reduction processing.
- the smart terminal further includes: The persistent link establishing unit is configured to establish a persistent voice link with the second terminal, and the microphone of the second terminal continuously sends the collected ambient noise to the first terminal.
- the embodiment of the invention provides a method for reducing noise of an intelligent terminal and an intelligent terminal, which collects call voice and environmental noise through the first terminal, and determines whether it is in a quiet state according to the collected call voice or/and environmental noise; If the terminal is not in a quiet state, the environment noise (which may include a slight voice sound wave signal) collected and transmitted by the second terminal is subjected to DSP (digital siganal process) denoising or noise reduction processing; In the quiet state, no noise reduction processing is performed, thereby effectively solving the problem that the smart terminal structure is small, and it is impossible to set a plurality of microphones for denoising processing, and the call is susceptible to environmental noise.
- DSP digital siganal process
- FIG. 1 is a schematic flow chart of a first embodiment of a method for reducing noise of a smart terminal according to the present invention
- FIG. 2 is a schematic flowchart of a second embodiment of a method for reducing noise of a smart terminal according to the present invention
- FIG. 4 is a schematic flowchart of a fourth embodiment of a method for reducing noise of a smart terminal according to the present invention
- FIG. 5 is a schematic flowchart of a fifth embodiment of a method for reducing noise of a smart terminal according to the present invention
- FIG. 7 is a schematic diagram of functional modules of a second embodiment of the intelligent terminal of the present invention
- FIG. 8 is a schematic diagram of functional modules of a third embodiment of the intelligent terminal of the present invention
- FIG. 10 is a schematic diagram of functional modules of a fifth embodiment of a smart terminal according to the present invention
- FIG. 11 is a schematic structural diagram of a specific implementation of the smart terminal of the present invention.
- FIG. 1 is a schematic flowchart diagram of a first embodiment of a noise reduction method of a smart terminal according to the present invention.
- the method for the voice loss of the smart terminal includes: Step S101: The first terminal collects the call voice and the ambient noise when the voice communication starts; Example 1. The microphone in the first terminal is used as the primary microphone. During the call, the first terminal collects the call voice and the ambient noise.
- the call voice signal and the ambient noise signal are collected through the main microphone, wherein the amplitude of the call voice signal is generally much larger than the environment.
- the amplitude of the noise signal is taken as the call voice signal by taking the maximum amplitude of the mixed voice signal of the collected call voice signal and the ambient noise signal; when the user of the first terminal is listening to the other party's speaking state, the ambient noise signal is collected through the main microphone, and then the acquisition is performed.
- the call voice signal and the ambient noise signal are transmitted to the processor of the first terminal.
- Example 2 The microphone in the first terminal is used as the main microphone. During the call, the first terminal collects the call voice and the ambient noise.
- the call voice signal is collected through the main microphone.
- the ambient noise signal wherein the amplitude of the call voice signal is generally much larger than the amplitude of the ambient noise signal, and the maximum amplitude of the mixed voice signal of the collected call voice signal and the ambient noise signal is taken as the call voice signal, and the collected call voice signal is transmitted to the a processor of the terminal; when the user of the first terminal is listening to the speaking state of the other party, collecting an environmental noise signal through the main microphone, and transmitting the collected environmental noise signal to the processor of the first terminal.
- the first terminal determines whether it is in a quiet state according to the call voice or/and the ambient noise collected by the first terminal.
- the manner in which the first terminal determines whether it is in a quiet state is different.
- the first terminal collects according to the first terminal.
- Example 2 The first terminal determines the maximum sound signal amplitude within a unit time (for example, 100 ms) of the collected ambient noise, and if it is determined that the maximum sound signal amplitude value of the ambient noise per unit time is less than a preset environmental noise amplitude threshold (for example, 40 dB), Then, it is determined that the first terminal is in a quiet state.
- a preset signal to noise ratio threshold for example, 6 decibels
- Step S103 if the first terminal is not in the quiet state, the first terminal performs noise reduction processing on the call voice according to the ambient noise sent by the second terminal; if the processor of the first terminal determines that it is not in the quiet state, that is, the first The terminal is in a relatively noisy environment, and the first terminal transmits the received ambient noise collected by the second terminal sub-microphone to the processor, wherein the second terminal performs A/D on the audio signal collected by the sub-microphone (A na).
- the log/digital conversion, that is, the analog-to-digital conversion, and the encoding processing, are transmitted to the first terminal through a wireless communication device, and the wireless communication device includes a short-range communication device such as Bluetooth, WIFI (Wireless Fidelity) near field communication.
- the processor of the first terminal synthesizes the call voice and ambient noise collected by the main microphone, and the ambient noise collected by the sub-microphone, and performs noise reduction processing on the call voice, for example, using a DSP (digital siganal process) algorithm to denoise or drop the call voice. Noise processing, finally encoding and outputting the denoised call speech to the base station, and then transmitting the data to the first terminal by the base station.
- DSP digital siganal process
- the DSP algorithm mostly adopts three general-purpose functions.
- the method of speech enhancement First, the interference subtraction method, that is, the noise is suppressed by subtracting the noise spectrum; secondly, the harmonic frequency suppression method, that is, the method of using speech enhancement to complete noise reduction, based on the periodic principle of noise, The adaptive comb filter of harmonic noise is used to implement fundamental frequency tracking to complete the noise reduction. Thirdly, the vocoder resynthesis method is used. It uses the iterative method to estimate the model parameters based on the speech modeling. The method of voice signal re-synthesizes the noiseless signal. Step S104: If the first terminal is in a quiet state, no noise reduction processing is performed.
- the processor of the first terminal determines that it is in a quiet state, that is, the first terminal is in a relatively quiet environment, the first terminal does not need to perform denoising processing on the call voice, and directly encodes and outputs the device to the device that communicates with the first terminal.
- the present invention provides a method for reducing noise of an intelligent terminal, which collects call voice and environmental noise through the first terminal, and determines whether it is in a quiet state according to the collected call voice or/and environmental noise; If the terminal is not in a quiet state, the environment noise (which may include a slight voice sound wave signal) collected and transmitted by the second terminal is subjected to DSP (digital siganal process) denoising or noise reduction processing; In the quiet state, noise reduction processing is not performed.
- the first terminal processor performs DSP denoising processing on the audio signal transmitted by the second terminal and the audio signal collected by the main microphone thereof.
- the main microphone and the sub-microphone are separated, the distance is long, and the main microphone of the first terminal during the call
- the main collection is the user's call voice and environmental noise, while the secondary microphone of the second terminal collects mainly the ambient noise.
- the difference between the signal and noise ratios of the audio signals collected by the two terminals is large, and can be better eliminated after being processed by the processor DSP algorithm.
- Ambient noise, reservation call voice No. Improves the quality of the call, thus effectively solving the problem that the smart terminal, especially the smart terminal, has a small structure, cannot set multiple microphones for denoising, and the call is susceptible to environmental noise.
- FIG. 2 is a schematic flowchart diagram of a second embodiment of a noise reduction method of a smart terminal according to the present invention.
- step S103 includes step S1031. If the first terminal is not in a quiet state, a voice link with the second terminal is established, and the microphone of the second terminal is started to collect ambient noise; if the first terminal does not In a quiet state, the wireless communication device of the first terminal establishes a voice link with the wireless communication device of the second terminal, and activates the secondary microphone of the second terminal to collect ambient noise.
- Step S1032 The first terminal performs noise reduction processing on the call voice according to the ambient noise sent by the second terminal.
- the second terminal converts and encodes the ambient noise A/D collected by the sub-microphone, and sends it to the first terminal through the wireless communication device, and the first terminal encodes and outputs the noise-reduced call speech.
- the first terminal selectively establishes a voice link with the second terminal according to the determination result of the quiet state, and if the first terminal is not in the quiet state, establishes a voice link with the second terminal, and starts the first
- the microphone of the second terminal collects the ambient noise, and performs noise reduction processing on the call voice, and can determine in real time whether the environment of the first terminal is in a quiet state, thereby determining whether to perform noise reduction processing on the call voice in real time.
- FIG. 3 is a schematic flowchart diagram of a third embodiment of a noise reduction method of a smart terminal according to the present invention.
- step S104 includes step S1041. If the first terminal is in a quiet state, the microphone of the second terminal is turned off to stop collecting ambient noise, and the voice link with the second terminal is disconnected; In the quiet state, the secondary microphone of the second terminal is remotely controlled to stop collecting ambient noise, and controls the wireless communication device to disconnect the voice link with the second terminal.
- Step S1042 The first terminal directly encodes and outputs the collected call voice. The first terminal does not perform noise reduction processing on the collected call voice, and directly outputs the collected call voice.
- the second terminal cooperates with the first terminal to close and open the voice link, so as to real-time determine whether the environment in which the terminal is located is in a quiet state, so as to determine whether to perform noise reduction processing on the call voice in real time.
- FIG. 4 is a schematic flowchart diagram of a fourth embodiment of a method for reducing noise of a smart terminal according to the present invention.
- the method further includes: Step S105: The first terminal establishes a persistent voice link with the second terminal, and the microphone of the second terminal continuously sends the collected ambient noise to the first terminal. After the first terminal collects the call voice and the ambient noise, the first terminal establishes a persistent voice link with the second terminal, and the voice link is no longer turned on and off, and the secondary microphone of the second terminal continuously collects the collected ambient noise.
- the first terminal sends to the first terminal, when the first terminal determines that it is in a quiet state, rejecting the ambient noise audio sent by the second terminal; when the first terminal determines that it is in an unquiet state, receiving the second terminal sending The ambient noise audio is then denoised for the call voice.
- the first terminal establishes a persistent voice link with the second terminal, and the first terminal does not perform the opening and closing of the voice link, which reduces the occupation time of the first terminal processor, and improves to a certain extent.
- the step S105 is after the step S101, and the step S105 can also be performed before the step S101, and the technical effects are exactly the same, and will not be described here.
- FIG. 5 is a schematic flowchart diagram of a fifth embodiment of a method for reducing noise of a smart terminal according to the present invention.
- the method further includes: Step S106: The first terminal monitors in real time whether the environment is in a quiet state, and determines whether the call voice needs to be noise-reduced; When the terminal is in an unquiet state, the call voice is subjected to noise reduction processing; if the first terminal is in a quiet state, the call voice is not subjected to noise reduction processing.
- the state does not perform noise reduction processing on the call voice, thereby ensuring the call quality of the call voice during the entire call.
- FIG. 6 is a schematic diagram of functional modules of a first embodiment of a smart terminal according to the present invention.
- the present invention further provides an intelligent terminal, comprising: a sound collecting unit 101, configured to collect call voice and ambient noise when the voice communication starts; Example 1, the microphone in the first terminal is used as the main microphone, during the call, the voice The collecting unit 101 collects the call voice and the ambient noise.
- the call voice signal and the ambient noise signal are collected by the main microphone, wherein the amplitude of the call voice signal is generally greater than the amplitude of the ambient noise signal.
- the ambient noise signal is collected by the sound collecting unit 101, and then the collected call is The voice signal and the ambient noise signal are transmitted to the processor of the first terminal.
- Example 2 The microphone in the first terminal is used as the main microphone. During the call, the sound collection unit 101 collects the call voice and the ambient noise. Preferably, when the user of the first terminal is in the speaking state, the call voice signal is collected through the main microphone.
- the ambient noise signal wherein the amplitude of the call voice signal is generally much larger than the amplitude of the ambient noise signal, and the maximum amplitude of the mixed voice signal of the collected call voice signal and the ambient noise signal is taken as the call voice signal, and the collected call voice signal is transmitted to
- the processor of the first terminal when the user of the first terminal is in the listening state of the other party, the ambient noise signal is collected by the sound collecting unit 101, and the collected environmental noise signal is transmitted to the processor of the first terminal.
- the quiet state determining unit 102 is configured to determine whether it is in a quiet state according to the call voice or/and the ambient noise collected by the primary microphone.
- the first terminal determines whether the state is in a quiet state, for example:
- the amplitude, the measurement unit is decibel
- a preset signal to noise ratio threshold for example, 6 decibels
- the quiet state determining unit 102 determines the maximum sound signal amplitude within a unit time (for example, 100 ms) of the collected ambient noise, and if it is determined that the maximum sound signal amplitude value per unit time of the ambient noise is less than a preset environmental noise amplitude threshold (for example, 40 dB) ), it is judged that the first terminal is in a quiet state.
- the voice noise reduction output unit 103 is configured to perform noise reduction processing on the call voice according to the ambient noise sent by the second terminal when the first terminal is not in the quiet state; if the quiet state determination unit 102 determines that it is not in the quiet state, The first terminal is in a relatively noisy environment, and the voice noise reduction output unit 103 transmits the received ambient noise collected by the second terminal sub-microphone to the processor, wherein the second terminal performs the audio signal collected by the sub-microphone.
- /IXAnalog/digital) conversion that is, analog-to-digital conversion, and encoding processing, and transmitted to the first terminal through a wireless communication device, which includes a short-range communication device such as Bluetooth, WIFI (Wireless Fidelity) near field communication.
- the processor of the voice noise reduction output unit 103 integrates the call voice and the ambient noise collected by the main microphone, and the ambient noise collected by the sub-microphone, and performs noise reduction processing on the call voice, for example, using a DSP (digital siganal process) algorithm to cancel the call voice. Noise or noise reduction processing, finally encoding and outputting the noise-reduced call voice to the device communicating with the first terminal.
- DSP digital siganal process
- DSP algorithms In order to reduce the noise of the signal during transmission and improve the quality of voice transmission, DSP algorithms mostly adopt three common methods of speech enhancement: First, the interference subtraction method, that is, suppressing noise by subtracting the noise spectrum; secondly, it is harmonic Wave frequency suppression method, that is, using voice enhancement method to complete noise reduction, based on the periodic principle of noise, adaptive noise filtering using harmonic noise to implement fundamental frequency tracking to complete noise reduction; Third, using vocoder The resynthesis method uses an iterative method to estimate model parameters based on speech modeling, and re-synthesizes the noiseless signal by describing the speech signal.
- the voice output unit 104 is configured to be in a quiet state at the first terminal, and does not perform noise reduction processing on the call voice.
- the quiet state determining unit 102 determines that the first terminal is in a quiet state, that is, the first terminal is in a relatively quiet environment, the first terminal does not need to perform denoising processing on the call voice, and the voice output unit 104 directly outputs the call voice code to the The device that the first terminal communicates can be.
- the present invention provides an intelligent terminal, which collects call voice and environmental noise through the sound collection unit 101, and the quiet state determination unit 102 determines whether it is in a quiet state according to the collected call voice or/and environmental noise;
- the terminal is not in a quiet state, and the voice noise reduction output unit 103 performs DSP (digital siganal process) denoising or noise reduction according to the ambient noise collected and transmitted by the second terminal (possibly including a slight voice sound wave signal).
- DSP digital siganal process
- the first terminal processor performs DSP denoising processing on the audio signal transmitted by the second terminal and the audio signal collected by the main microphone thereof. Since the main microphone and the sub-microphone are separated, the distance is long, and the first terminal of the call is The main microphone is mainly collected by the user's call voice and ambient noise, while the secondary microphone of the second terminal is mainly collected by the ambient noise. The signal-to-noise ratio of the audio signals collected by the two is quite different, which can be better processed by the processor DSP algorithm. The elimination of environmental noise, the preservation of the call voice signal, and the improvement of the call quality, thereby effectively solving the problem that the smart terminal, especially the smart terminal, is small in structure, cannot set multiple microphones for denoising, and the call is susceptible to environmental noise.
- FIG. 7 is a schematic diagram of functional modules of a second embodiment of the intelligent terminal of the present invention.
- the voice noise reduction output unit 103 of the smart terminal includes: a link enabling module 1031, configured to establish a voice link with the second terminal when the first terminal is not in a quiet state, and start a microphone of the second terminal to collect ambient noise; The first terminal is not in a quiet state, and the link enabling module 1031 establishes a voice link with the wireless communication device of the second terminal, and starts the secondary microphone of the second terminal to collect ambient noise.
- the noise reduction output module 1032 is configured to perform noise reduction processing according to ambient noise sent by the second terminal.
- the second terminal converts and encodes the ambient noise A/D collected by the sub-microphone, and sends it to the first terminal through the wireless communication device, and the module 1032 performs noise reduction processing according to the ambient noise sent by the second terminal.
- the link enabling module 1031 selectively establishes a voice link with the second terminal according to the determination result of the quiet state of the first terminal, and establishes a second terminal with the second terminal if the first terminal is not in the quiet state.
- the voice link starts the microphone of the second terminal to collect the ambient noise
- the noise reduction output module 1032 performs noise reduction processing on the call voice, and then encodes and outputs the noise reduced call voice, and can determine in real time whether the environment of the first terminal is in a quiet state. , thereby determining whether to perform noise reduction processing on the call voice in real time.
- FIG. 8 is a schematic diagram of functional modules of a third embodiment of the intelligent terminal of the present invention.
- the voice output unit 104 of the smart terminal includes: a link close module 1041, configured to: when the first terminal is in a quiet state, turn off the microphone of the second terminal to stop collecting ambient noise, and disconnect the voice link with the second terminal; When the terminal is in a quiet state, the link closing module 1041 remotely controls the secondary microphone of the second terminal to stop collecting ambient noise, and controls the wireless communication device to disconnect the voice link with the second terminal.
- the direct output module 1042 is configured to directly encode and output the collected call voice.
- the first terminal does not perform noise reduction processing on the collected call voice, and the direct output module 1042 directly programs and outputs the collected call voice.
- the second terminal cooperates with the first terminal to close and open the voice link, so as to determine whether the environment in which the first terminal is located is in a quiet state in real time, so as to determine whether to perform noise reduction processing on the call voice in real time.
- FIG. 9 is a schematic diagram of functional modules of a fourth embodiment of the intelligent terminal of the present invention.
- the smart terminal further includes: a persistent link establishing unit 105, configured to establish a persistent voice link with the second terminal, and the microphone of the second terminal continuously sends the collected ambient noise to the first terminal to collect the call voice in the sound collecting unit 101.
- the persistent link establishing unit 105 establishes a persistent voice link with the second terminal, and no longer performs the opening and closing of the voice link, and the secondary microphone of the second terminal continuously sends the collected ambient noise to the first The terminal, when the first terminal determines that it is in a quiet state, rejects the ambient noise audio sent by the second terminal; when the first terminal determines that it is in an unquiet state, receives the ambient noise audio sent by the second terminal, Then, the call voice is denoised.
- the first terminal establishes a persistent voice link with the second terminal, and the first terminal does not perform the opening and closing of the voice link, which reduces the occupation time of the first terminal processor, and improves to a certain extent. The efficiency of noise reduction processing on call voice.
- FIG. 10 is a schematic diagram of functional modules of a fifth embodiment of the intelligent terminal of the present invention.
- the smart terminal further includes: a real-time monitoring unit 106, configured to monitor in real time whether the environment in which the first terminal is located is in a quiet state, determine whether the call voice needs to be noise-reduced; if the first terminal is in an unquiet state, The call voice performs noise reduction processing; if the first terminal is in a quiet state, the call voice is not subjected to noise reduction processing.
- a real-time monitoring unit 106 configured to monitor in real time whether the environment in which the first terminal is located is in a quiet state, determine whether the call voice needs to be noise-reduced; if the first terminal is in an unquiet state, The call voice performs noise reduction processing; if the first terminal is in a quiet state, the call voice is not subjected to noise reduction processing.
- whether the real-time monitoring unit 106 determines whether the environment in which the first terminal is located is in a quiet state, whether to perform noise reduction processing on the call voice, and continuously monitors whether the current environment is in a quiet state during subsequent calls. And according to the environmental condition of the subsequent real-time monitoring, it is judged whether the noise reduction process or the noise reduction process is performed, thereby realizing the correct processing of the call voice in real time during the call, that is, the noise reduction of the call voice is performed in the non-quiet state. In the quiet state, the voice of the call is not subjected to noise reduction processing, thereby ensuring the quality of the call voice during the entire call.
- FIG. 11 is a schematic structural diagram of a specific implementation of the smart terminal of the present invention.
- the second terminal comprises a microphone, a processor, a wireless communication device and a power source, the microphone of the second terminal serves as a secondary microphone, the first terminal comprises a microphone, a processor, a wireless communication device and a power source, the microphone of the first terminal serves as a main microphone, and the wireless communication
- the device includes a short-range communication device such as Bluetooth and WIFI. Taking Bluetooth as an embodiment, the Bluetooth communication device includes a Bluetooth chip and an antenna. The first terminal establishes a Bluetooth pairing with the second terminal.
- the processor When the first terminal detects the noise environment, the processor encodes the built-in control command to the Bluetooth chip, and the Bluetooth chip encodes the command string according to the Bluetooth ATX Attention) instruction set. Transmitting to the second terminal Bluetooth communication device via the Bluetooth antenna, the second terminal Bluetooth communication device converts the Bluetooth AT command transmitted by the first terminal into a communication standard AT command, and sends the signal to the second terminal processor, and the second terminal processor The AT command is executed to establish a Bluetooth voice link between the first terminal and the second terminal, and the secondary microphone is started to collect the audio signal, and then A/D converted and encoded, and then sent to the first terminal Bluetooth communication device through the Bluetooth communication device.
- the control unit sends a control command to the second terminal to turn off the secondary microphone, and disconnects the Bluetooth voice link.
- the first terminal Bluetooth communication device and the second terminal Bluetooth communication device establish a voice link connection, and the second terminal sub-microphone always collects ambient noise and sends the same to the first terminal, and the first terminal detects the call environment. Determine if dual microphone denoising is performed.
- the microphone of the first terminal can be used as a secondary microphone during the call, and is used to collect the call environment noise and send it to the second terminal.
- the microphone of the second terminal is used as the main microphone for collecting voice and environmental noise during the call, and the second terminal performs DSP denoising processing on the audio signal transmitted by the first terminal processor and the audio signal collected by the microphone itself.
- the two microphones are separated, the distance is long, and the amplitude of the collected speech signals is large. After the processor DSP is denoised, the good speech signal can be retained.
- This embodiment can effectively solve the problem that the existing dual microphone noise canceling mobile phone is free due to the limited distance of the two microphones. Sending loudness when talking is affected by the distance between the mouth and the phone.
- the method for the call noise reduction according to the present invention is basically the same as the above, and the method is basically the same as the method for the first terminal to reduce the noise, and the difference is: in step S103 The coded output denoised call voice is sent to the second terminal, and then the second terminal sends the denoised call voice to the device that communicates with the first terminal; in step S104, the first terminal directly encodes The collected call voice is outputted to the second terminal, and then the second terminal sends the noise-reduced call voice to the device communicating with the first terminal.
- the first terminal and the second terminal may be used as a processing unit for performing noise reduction processing on the call voice and the ambient noise collected by the first terminal.
- the first terminal collects The voice call and the ambient noise are directly sent to the processor of the first terminal, and the processor of the first terminal receives the ambient noise collected by the second terminal, and then the first terminal performs noise reduction on the call voice according to the ambient noise sent by the second terminal.
- the second terminal And encoding and outputting the noise-reduced call voice to the second terminal, and then the second terminal sends the noise-reduced call voice to the device that communicates with the first terminal; when the second terminal acts as the noise reduction processor, the first The terminal collects the voice call and the ambient noise and sends the voice to the processor of the second terminal, and then the second terminal performs the noise reduction processing on the call voice according to the call voice and the ambient noise sent by the first terminal, and the second terminal will reduce the voice of the call after the noise reduction. Send to the device that communicates with the first terminal.
- the above is only a preferred embodiment of the present invention, and is not intended to limit the scope of the present invention, and equivalent structural or process changes made by the present specification and drawings may be directly or indirectly applied to other related technical fields.
- the technical solution provided by the foregoing embodiments of the present invention can be applied to the denoising processing of the intelligent terminal in the communication field, and is particularly applicable to the denoising processing of the wearable intelligent terminal, and the smart solution of the technical solution provided by the above embodiment of the present invention is applied.
- the terminal can effectively eliminate noise and improve call quality.
Landscapes
- Engineering & Computer Science (AREA)
- Signal Processing (AREA)
- Telephone Function (AREA)
Abstract
本发明公开了一种智能终端通话降噪的方法，包括：在语音通信开始时，第一终端采集通话语音和环境噪声；第一终端根据其采集的通话语音或/和环境噪声，判断是否处于安静状态；若第一终端不处于安静状态，则根据第二终端发送过来的环境噪声对通话语音进行降噪处理；若第一终端处于安静状态，则不进行降噪处理。本发明同时公开一种智能终端。本发明有效解决了因智能终端，特别是穿戴智能终端结构小巧，无法设置多个麦克风进行消噪处理，通话易受环境噪音干扰的问题。
Description
智能终端通话降噪方法及智能终端 技术领域 本发明涉及通信技术领域， 尤其涉及一种智能终端的降噪方法及智能终端。 背景技术 智能终端， 特别是穿戴智能终端 （例如， 智能手表、 智能眼镜等） 作为一种新型 的智能终端， 可方便地穿戴在人们的特定身体部位。 现有的穿戴智能终端一般都具备 了无线通话功能， 其通常被用于特殊的环境之下， 例如， 当人们的双手无法持有或者 操作移动终端 （例如， 手机、 平板电脑等） 时， 通过使用这类穿戴智能终端与直接与 其他终端设备进行通信， 大大扩展了人们语音通话的场景。 但是， 受限于穿戴智能终 端一般结构比较小巧， 无法设置由多个 MIC (Microphone, 麦克风） 组成的阵列进行 消噪处理， 因此， 这类穿戴智能终端的通话通常易受环境噪音干扰， 尤其是在比较嘈 杂的环境下， 通话质量不能得到有效保证。 上述内容仅用于辅助理解本发明的技术方案，并不代表承认上述内容是现有技术。 发明内容 本发明实施例提供了一种智能终端的降噪方法及智能终端， 旨在解决现有智能终 端， 特别是针对穿戴智能终端因体积小无法进行消噪处理的问题。 根据本发明的一个方面， 提供了一种智能终端通话降噪的方法， 包括： 在语音通信开始时， 第一终端采集通话语音和环境噪声； 第一终端根据其采集的通话语音或 /和环境噪声， 判断是否处于安静状态； 若所述第一终端不处于安静状态， 则所述第一终端根据所述第二终端发送过来的 环境噪声对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不做降噪处理。 优选地， 所述第一终端根据其采集的通话语音或 /和环境噪声， 判断是否处于安静 状态的步骤包括：
根据所述第一终端采集的通话语音和环境噪声得出信噪比； 若所述信噪比大于等于预设值， 则所述第一终端处于安静状态。 优选地， 所述第一终端根据其采集的通话语音或 /和环境噪声， 判断是否处于安静 状态的步骤包括： 根据所述第一终端采集的环境噪声得出环境噪声的幅度； 若所述环境噪声幅度值小于预设噪声信号幅度阀值， 则所述第一终端处于安静状 态。 优选地， 所述若第一终端不处于安静状态， 则所述第一终端根据所述第二终端发 送过来的环境噪声对通话语音进行降噪处理的步骤包括： 若所述第一终端不处于安静状态， 则建立与所述第二终端的语音链路， 启动所述 第二终端的麦克风采集环境噪声； 所述第一终端根据所述第二终端发送过来的环境噪声进行降噪处理。 优选地， 所述第一终端不处于安静状态进行降噪处理， 或者所述第一终端处于安 静状态不进行降噪处理的步骤之后还包括： 所述第一终端实时监测其所处环境是否是安静状态， 判断是否需要对通话语音进 行降噪； 若所述第一终端处于不安静状态， 则对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不对通话语音进行降噪处理。 优选地， 在所述第一终端的主麦克风采集通话语音和环境噪声的步骤之后， 还包 括： 所述第一终端建立与所述第二终端的持久语音链路， 所述第二终端的麦克风不断 将采集到的环境噪声发送给所述第一终端。
根据本发明的另一方面， 提供了一种智能终端， 包括： 声音采集单元， 设置为在语音通信开始时， 采集通话语音和环境噪声；
安静状态判断单元， 设置为根据所述主麦克风采集的通话语音或 /和环境噪声， 判 断是否处于安静状态； 语音降噪输出单元， 设置为在所述第一终端不处于安静状态时， 根据所述第二终 端发送过来的环境噪声对通话语音进行降噪处理，； 语音输出单元， 设置为在所述第一终端处于安静状态，不对通话语音做降噪处理。 优选地， 所述安静状态判断单元设置为： 根据所述第一终端采集的通话语音和环境噪声得出信噪比； 若所述信噪比大于等于预设值， 则所述第一终端处于安静状态 优选地， 所述安静状态判断单元设置为： 根据所述第一终端采集的环境噪声得出环境噪声的幅度； 若所述环境噪声幅度值小于预设噪声信号幅度阀值， 则所述第一终端处于安静状 态。 优选地， 所述语音降噪输出单元包括： 链路启用模块， 设置为在所述第一终端不处于安静状态时， 建立与所述第二终端 的语音链路， 启动所述第二终端的麦克风采集环境噪声； 降噪输出模块， 设置为根据所述第二终端发送过来的环境噪声对通话语音进行降 噪处理。 优选地， 所述智能终端还包括： 实时监测单元， 设置为实时监测所述第一终端所处环境是否是安静状态， 判断是 否需要对通话语音进行降噪； 若所述第一终端处于不安静状态， 则对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不对通话语音进行降噪处理。 优选地， 所述智能终端还包括：
持久链路建立单元， 设置为建立与所述第二终端的持久语音链路， 所述第二终端 的麦克风不断将采集到的环境噪声发送给所述第一终端。
本发明实施例提供了一种智能终端的降噪方法及智能终端， 通过第一终端采集通 话语音和环境噪声， 并根据采集的通话语音或 /和环境噪声， 判断是否处于安静状态； 若第一终端不处于安静状态， 则根据第二终端采集并发送过来的环境噪声 （有可能包 括轻微的语音声波信号） 对通话语音进行 DSP (digital siganal process) 消噪或降噪处 理； 若第一终端处于安静状态， 则不做降噪处理， 从而有效解决了因智能终端结构小 巧， 无法设置多个麦克风进行消噪处理， 通话易受环境噪音干扰的问题。 附图说明 图 1是本发明智能终端的降噪方法第一实施例的流程示意图； 图 2是本发明智能终端的降噪方法第二实施例的流程示意图； 图 3是本发明智能终端的降噪方法第三实施例的流程示意图； 图 4是本发明智能终端的降噪方法第四实施例的流程示意图； 图 5是本发明智能终端的降噪方法第五实施例的流程示意图； 图 6是本发明智能终端第一实施例的功能模块示意图； 图 7是本发明智能终端第二实施例的功能模块示意图； 图 8是本发明智能终端第三实施例的功能模块示意图； 图 9是本发明智能终端第四实施例的功能模块示意图； 图 10是本发明智能终端第五实施例的功能模块示意图； 以及 图 11是本发明智能终端具体实现结构示意图。
为了使本发明的技术方案更加清楚、 明了， 下面将结合附图作进一步详述。
具体实施方式 应当理解，此处所描述的具体实施例仅仅用以解释本发明， 并不用于限定本发明。 本发明提供一种智能终端通话降噪的方法。 参见图 1， 图 1是本发明智能终端的降噪方法第一实施例的流程示意图。 在第一实施例中， 智能终端通话降噪的方法， 包括： 步骤 S101 , 在语音通信开始时， 第一终端采集通话语音和环境噪声； 例一、 第一终端内的麦克风作为主麦克风， 在通话过程中， 第一终端采集通话语 音和环境噪声， 优选地， 在第一终端的用户处于说话状态时， 通过主麦克风采集通话 语音信号和环境噪声信号，其中通话语音信号的幅度一般远大于环境噪声信号的幅度， 取所采集通话语音信号和环境噪声信号混合声音信号的最大幅度作为通话语音信号； 在第一终端的用户处于听对方说话状态时， 通过主麦克风采集环境噪声信号， 然后将 采集的通话语音信号和环境噪声信号传送至第一终端的处理器。 例二、 第一终端内的麦克风作为主麦克风， 在通话过程中， 第一终端采集通话语 音和环境噪声， 优选地， 在第一终端的用户处于说话状态时， 通过主麦克风采集通话 语音信号和环境噪声信号，其中通话语音信号的幅度一般远大于环境噪声信号的幅度， 取所采集通话语音信号和环境噪声信号混合声音信号的最大幅度作为通话语音信号， 并将采集的通话语音信号传送至第一终端的处理器； 在第一终端的用户处于听对方说 话状态时， 通过主麦克风采集环境噪声信号， 并将采集的环境噪声信号传送至第一终 端的处理器。 步骤 S102,第一终端根据其采集的通话语音或 /和环境噪声，判断是否处于安静状 态； 第一终端判断是否处于安静状态的方式有多种， 例如： 例一、 第一终端根据其采集的通话语音信号的最大幅度与环境噪声信号的最大幅 度相比得出信噪比 （信噪比 =10LG(Ps/Pn)， 其中 Ps和 Pn分别代表信号和噪声的有效 功率或幅度， 计量单位为分贝），若主麦克风采集的通话语音信噪比大于等于预设信噪 比阀值 （例如， 6分贝）， 则判断第一终端处于安静状态；
例二、 第一终端确定采集的环境噪声的单位时间 （例如 100ms) 内最大声音信号 幅度，若确定环境噪声单位时间内最大声音信号幅度值小于预设环境噪声幅度阀值（例 如 40分贝）， 则判断第一终端处于安静状态。 步骤 S103 , 若第一终端不处于安静状态， 则第一终端根据第二终端发送过来的环 境噪声对通话语音进行降噪处理； 若第一终端的处理器判定其不处于安静状态，即第一终端处于较为嘈杂的环境下， 第一终端则将接收的第二终端副麦克风采集的环境噪声传至处理器， 其中， 第二终端 对其副麦克风采集到的音频信号进行 A/D(Analog/digital)转换， 即模数转换， 和编码处 理， 并通过无线通讯装置发送至第一终端， 该无线通讯装置包括蓝牙、 WIFI (Wireless Fidelity ) 近场通信等近距离通讯装置。 第一终端的处理器综合主麦克风采集的通话 语音和环境噪声， 及副麦克风采集的环境噪声， 对通话语音进行降噪处理， 例如采用 DSP (digital siganal process) 算法对通话语音进行消噪或降噪处理， 最后编码输出降 噪后的通话语音至基站， 然后通过基站传输给与第一终端通话的设备 为了降低信号在传输过程中的噪声， 改善语音传输质量， DSP算法大多会采用三 种通用的语音增强的方法： 首先， 是干扰相减法， 即通过减掉噪声频谱来抑制噪声； 其次， 是谐波频率抑制法， 即利用语音增强的方法来完成减噪， 基于噪声的周期性原 理， 利用谐波噪声的自适应梳状滤波实施基频跟踪来完成降噪； 第三， 是利用声码器 再合成法， 它利用迭代法， 在语音建模的基础上， 估计模型参数， 用描述语音信号的 方法再重新合成无噪声信号。 步骤 S104, 若第一终端处于安静状态， 则不进行降噪处理。 若第一终端的处理器判定其处于安静状态， 即第一终端处于较为安静的环境下， 第一终端无需对通话语音进行消噪处理，直接编码输出至与第一终端通信的设备即可。 在本实施例中， 本发明提出一种智能终端的降噪方法， 通过第一终端采集通话语 音和环境噪声， 并根据采集的通话语音或 /和环境噪声， 判断是否处于安静状态； 若第 一终端不处于安静状态， 则根据第二终端采集并发送过来的环境噪声 （有可能包括轻 微的语音声波信号） 对通话语音进行 DSP (digital siganal process) 消噪或降噪处理； 若第一终端处于安静状态， 则不进行降噪处理。 这样第一终端处理器对第二终端传送 过来的音频信号和其自身的主麦克风采集到音频信号进行 DSP消噪处理， 由于主麦克 风和副麦克风分离， 距离远， 通话时第一终端的主麦克风采集的主要是用户通话语音 和环境噪音， 而第二终端的副麦克风采集的主要是环境噪音， 两者采集的音频信号信 噪比差异较大， 经处理器 DSP算法处理后可以较好的消除环境噪声， 保留通话语音信
号， 提高通话质量， 从而有效解决了因智能终端， 特别是穿戴智能终端结构小巧， 无 法设置多个麦克风进行消噪处理， 通话易受环境噪音干扰的问题。
参见图 2， 图 2是本发明智能终端的降噪方法第二实施例的流程示意图。 在第一实施例的基础上， 步骤 S103包括 步骤 S1031 , 若第一终端不处于安静状态， 则建立与第二终端的语音链路， 启动 第二终端的麦克风采集环境噪声； 若第一终端不处于安静状态， 则第一终端的无线通讯装置与第二终端的无线通讯 装置建立语音链路， 并启动第二终端的副麦克风采集环境噪声。 步骤 S1032, 第一终端根据第二终端发送过来的环境噪声对通话语音进行降噪处 理。 第二终端将其副麦克风采集的环境噪声 A/D转换和编码， 并通过无线通讯装置发 送给第一终端， 第一终端编码输出降噪后的通话语音。 在本实施例中， 第一终端根据安静状态的判定结果， 有选择地与第二终端建立语 音链路， 若第一终端不处于安静状态， 则建立与第二终端的语音链路， 启动第二终端 的麦克风采集环境噪声， 并对通话语音进行降噪处理， 可以实时判定第一终端所处环 境是否处于安静状态， 从而实时判定是否对通话语音进行降噪处理。
参见图 3， 图 3是本发明智能终端的降噪方法第三实施例的流程示意图。 在第二实施例的基础上， 步骤 S104包括 步骤 S1041 , 若第一终端处于安静状态， 则关闭第二终端的麦克风停止采集环境 噪声， 断开与第二终端的语音链路； 若第一终端处于安静状态， 则远程控制第二终端的副麦克风停止采集环境噪声， 并控制其无线通讯装置断开与第二终端的语音链路。 步骤 S1042, 第一终端直接编码输出其采集的通话语音。 第一终端不对采集的通话语音进行降噪处理， 直接编程输出采集的通话语音。
在本实施例中， 第二终端配合第一终端关闭和开启语音链路， 以实现实时判定第 终端所处环境是否处于安静状态， 从而实时判定是否对通话语音进行降噪处理。
参见图 4， 图 4是本发明智能终端的降噪方法第四实施例的流程示意图。 在第一实施例的基础上， 步骤 S101之后还包括： 步骤 S105 , 第一终端建立与第二终端的持久语音链路， 第二终端的麦克风不断将 采集到的环境噪声发送给第一终端。 在第一终端采集通话语音和环境噪声之后， 第一终端建立与第二终端的持久语音 链路， 不再进行语音链路的开启和关闭， 第二终端的副麦克风不断将采集到的环境噪 声发送给第一终端， 当第一终端判定其处于安静状态下， 则拒收第二终端发送过来的 环境噪声音频； 当第一终端判定其处于不安静状态下， 则接收第二终端发送过来的环 境噪声音频， 再对通话语音进行降噪处理。 在本实施例中， 第一终端建立与第二终端的持久语音链路， 第一终端不再进行语 音链路的开启和关闭， 减少了第一终端处理器的占用时间， 在一定程度上提高了对通 话语音进行降噪处理的效率。 另外， 除了在本实施例中， 步骤 S105处于步骤 S101之后， 步骤 S105也可以在 步骤 S101之前， 所达到的技术效果完全一样， 在此不做累述。
参见图 5， 图 5是本发明智能终端的降噪方法第五实施例的流程示意图。 在第一实施例的基础上， 步骤 S103或者步骤 S104之后还包括： 步骤 S106, 第一终端实时监测其所处环境是否是安静状态， 判断是否需要对通话 语音进行降噪； 若所述第一终端处于不安静状态， 则对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不对通话语音进行降噪处理。 在本实施例中， 无论第一终端对其所处的环境判断是否为安静状态， 是否对通话 语音进行降噪处理， 在后续的通话过程中都继续监测当前的环境是否为安静状态， 并
根据后续实时监测的环境情况判断是进入降噪处理还是退出降噪处理， 从而实现了在 通话过程中实时对通话语音进行正确处理， 即在不安静状态则对通话语音进行降噪处 理， 在安静状态则不对通话语音进行降噪处理， 进而保证通话语音在整个通话过程中 的通话质量。
参见图 6， 图 6是本发明智能终端第一实施例的功能模块示意图。 本发明进一步提出一种智能终端， 包括： 声音采集单元 101， 设置为在语音通信开始时， 采集通话语音和环境噪声； 例一、 第一终端内的麦克风作为主麦克风， 在通话过程中， 声音采集单元 101采 集通话语音和环境噪声， 优选地， 在第一终端的用户处于说话状态时， 通过主麦克风 采集通话语音信号和环境噪声信号， 其中通话语音信号的幅度一般远大于环境噪声信 号的幅度， 取所采集通话语音信号和环境噪声信号混合声音信号的最大幅度作为通话 语音信号； 在第一终端的用户处于听对方说话状态时， 通过声音采集单元 101采集环 境噪声信号， 然后将采集的通话语音信号和环境噪声信号传送至第一终端的处理器。 例二、 第一终端内的麦克风作为主麦克风， 在通话过程中， 声音采集单元 101采 集通话语音和环境噪声， 优选地， 在第一终端的用户处于说话状态时， 通过主麦克风 采集通话语音信号和环境噪声信号， 其中通话语音信号的幅度一般远大于环境噪声信 号的幅度， 取所采集通话语音信号和环境噪声信号混合声音信号的最大幅度作为通话 语音信号， 并将采集的通话语音信号传送至第一终端的处理器； 在第一终端的用户处 于听对方说话状态时， 通过声音采集单元 101采集环境噪声信号， 并将采集的环境噪 声信号传送至第一终端的处理器。 安静状态判断单元 102， 设置为根据主麦克风采集的通话语音或 /和环境噪声， 判 断是否处于安静状态； 第一终端判断是否处于安静状态的方式有多种， 例如： 例一、 安静状态判断单元 102根据其采集的通话语音信号的最大幅度与环境噪声 信号的最大幅度相比得出信噪比 （信噪比 =10LG(Ps/Pn)， 其中 Ps和 Pn分别代表信号 和噪声的有效功率或幅度， 计量单位为分贝），若主麦克风采集的通话语音信噪比大于 等于预设信噪比阀值 （例如， 6分贝）， 则判断第一终端处于安静状态；
例二、 安静状态判断单元 102确定采集的环境噪声的单位时间 （例如 100ms) 内 最大声音信号幅度， 若确定环境噪声单位时间内最大声音信号幅度值小于预设环境噪 声幅度阀值 （例如 40分贝）， 则判断第一终端处于安静状态。 语音降噪输出单元 103， 设置为在第一终端不处于安静状态时， 根据第二终端发 送过来的环境噪声对通话语音进行降噪处理； 若安静状态判断单元 102判定其不处于安静状态， 即第一终端处于较为嘈杂的环 境下， 语音降噪输出单元 103则将接收的第二终端副麦克风采集的环境噪声传至处理 器， 其中， 第二终端对其副麦克风采集到的音频信号进行 A/IXAnalog/digital)转换， 即 模数转换， 和编码处理， 并通过无线通讯装置发送至第一终端， 该无线通讯装置包括 蓝牙、 WIFI (Wireless Fidelity ) 近场通信等近距离通讯装置。 语音降噪输出单元 103 的处理器综合主麦克风采集的通话语音和环境噪声， 及副麦克风采集的环境噪声， 对 通话语音进行降噪处理， 例如采用 DSP (digital siganal process) 算法对通话语音进行 消噪或降噪处理， 最后编码输出降噪后的通话语音至与第一终端通信的设备。 为了降低信号在传输过程中的噪声， 改善语音传输质量， DSP算法大多会采用三 种通用的语音增强的方法： 首先， 是干扰相减法， 即通过减掉噪声频谱来抑制噪声； 其次， 是谐波频率抑制法， 即利用语音增强的方法来完成减噪， 基于噪声的周期性原 理， 利用谐波噪声的自适应梳状滤波实施基频跟踪来完成降噪； 第三， 是利用声码器 再合成法， 它利用迭代法， 在语音建模的基础上， 估计模型参数， 用描述语音信号的 方法再重新合成无噪声信号。 语音输出单元 104， 设置为在第一终端处于安静状态， 不对通话语音做降噪处理。 若安静状态判断单元 102器判定第一终端处于安静状态， 即第一终端处于较为安 静的环境下， 第一终端无需对通话语音进行消噪处理， 语音输出单元 104直接将通话 语音编码输出至与第一终端通信的设备即可。 在本实施例中， 本发明提出一种智能终端， 通过声音采集单元 101采集通话语音 和环境噪声， 安静状态判断单元 102根据采集的通话语音或 /和环境噪声， 判断是否处 于安静状态； 若第一终端不处于安静状态， 语音降噪输出单元 103则根据第二终端采 集并发送过来的环境噪声 （有可能包括轻微的语音声波信号） 对通话语音进行 DSP (digital siganal process) 消噪或降噪处理， 编码输出降噪后的通话语音； 若第一终端 处于安静状态，语音输出单元 104则直接编码输出声音采集单元 101采集的通话语音。 这样第一终端处理器对第二终端传送过来的音频信号和其自身的主麦克风采集到音频 信号进行 DSP消噪处理， 由于主麦克风和副麦克风分离， 距离远， 通话时第一终端的
主麦克风采集的主要是用户通话语音和环境噪音， 而第二终端的副麦克风采集的主要 是环境噪音， 两者采集的音频信号信噪比差异较大， 经处理器 DSP算法处理后可以较 好的消除环境噪声， 保留通话语音信号， 提高通话质量， 从而有效解决了因智能终端， 特别是穿戴智能终端结构小巧， 无法设置多个麦克风进行消噪处理， 通话易受环境噪 音干扰的问题。
参见图 7， 图 7是本发明智能终端第二实施例的功能模块示意图。 智能终端的语音降噪输出单元 103包括： 链路启用模块 1031， 设置为在第一终端不处于安静状态时， 建立与第二终端的语 音链路， 启动第二终端的麦克风采集环境噪声； 若第一终端不处于安静状态，则链路启用模块 1031与第二终端的无线通讯装置建 立语音链路， 并启动第二终端的副麦克风采集环境噪声。 降噪输出模块 1032， 设置为根据第二终端发送过来的环境噪声进行降噪处理。 第二终端将其副麦克风采集的环境噪声 A/D转换和编码， 并通过无线通讯装置发 送给第一终端， 模块 1032根据第二终端发送过来的环境噪声进行降噪处理。 在本实施例中，链路启用模块 1031根据第一终端的安静状态的判定结果，有选择 地与第二终端建立语音链路， 若第一终端不处于安静状态， 则建立与第二终端的语音 链路， 启动第二终端的麦克风采集环境噪声， 降噪输出模块 1032对通话语音进行降噪 处理， 再编码输出降噪后的通话语音， 可以实时判定第一终端所处环境是否处于安静 状态， 从而实时判定是否对通话语音进行降噪处理。
参见图 8， 图 8是本发明智能终端第三实施例的功能模块示意图。 智能终端的语音输出单元 104包括： 链路关闭模块 1041， 设置为在第一终端处于安静状态时， 关闭第二终端的麦克风 停止采集环境噪声， 断开与第二终端的语音链路； 若第一终端处于安静状态，链路关闭模块 1041则远程控制第二终端的副麦克风停 止采集环境噪声， 并控制其无线通讯装置断开与第二终端的语音链路。
直接输出模块 1042， 设置为直接编码输出其采集的通话语音。 第一终端不对采集的通话语音进行降噪处理，直接输出模块 1042则直接编程输出 采集的通话语音。 在本实施例中， 第二终端配合第一终端关闭和开启语音链路， 以实现实时判定第 一终端所处环境是否处于安静状态， 从而实时判定是否对通话语音进行降噪处理。
参见图 9， 图 9是本发明智能终端第四实施例的功能模块示意图。 智能终端还包括： 持久链路建立单元 105， 设置为建立与第二终端的持久语音链路， 第二终端的麦 克风不断将采集到的环境噪声发送给第一终端 在声音采集单元 101采集通话语音和环境噪声之后， 持久链路建立单元 105建立 与第二终端的持久语音链路， 不再进行语音链路的开启和关闭， 第二终端的副麦克风 不断将采集到的环境噪声发送给第一终端， 当第一终端判定其处于安静状态下， 则拒 收第二终端发送过来的环境噪声音频； 当第一终端判定其处于不安静状态下， 则接收 第二终端发送过来的环境噪声音频， 再对通话语音进行降噪处理。 在本实施例中， 第一终端建立与第二终端的持久语音链路， 第一终端不再进行语 音链路的开启和关闭， 减少了第一终端处理器的占用时间， 在一定程度上提高了对通 话语音进行降噪处理的效率。
参见图 10， 图 10是本发明智能终端第五实施例的功能模块示意图。 智能终端还包括： 实时监测单元 106， 设置为实时监测所述第一终端所处环境是否是安静状态， 判 断是否需要对通话语音进行降噪； 若所述第一终端处于不安静状态， 则对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不对通话语音进行降噪处理。
在本实施例中， 无论实时监测单元 106对第一终端所处的环境判断是否为安静状 态， 是否对通话语音进行降噪处理， 在后续的通话过程中都继续监测当前的环境是否 为安静状态，并根据后续实时监测的环境情况判断是进入降噪处理还是退出降噪处理， 从而实现了在通话过程中实时对通话语音进行正确处理， 即在不安静状态则对通话语 音进行降噪处理， 在安静状态则不对通话语音进行降噪处理， 进而保证通话语音在整 个通话过程中的通话质量。
参见图 11， 图 11是本发明智能终端具体实现结构示意图。 第二终端包括麦克风、 处理器、 无线通讯装置和电源， 第二终端的麦克风作为副 麦克风， 第一终端包括麦克风、 处理器、 无线通讯装置和电源， 第一终端的麦克风作 为主麦克风， 无线通讯装置包括蓝牙、 WIFI等近距离通讯装置。 以蓝牙为实施例， 蓝 牙通讯装置包括蓝牙芯片和天线。 第一终端与第二终端建立蓝牙配对， 第一终端检测 到噪声环境时， 其处理器将内置的控制命令进行编码发送给蓝牙芯片， 蓝牙芯片对该 命令字符串按蓝牙 ATXAttention)指令集合进行编码， 经蓝牙天线发送给第二终端蓝牙 通讯装置， 第二终端蓝牙通讯装置将第一终端传送过来的蓝牙 AT指令转换为通讯标 准 AT指令， 并发送给第二终端处理器， 第二终端处理器执行 AT指令， 建立第一终端 备和第二终端的蓝牙语音链路， 启动副麦克风采集音频信号， 进行 A/D转换和编码后 通过蓝牙通讯装置发送给第一终端蓝牙通讯装置。 当第一终端检测到安静环境时， 按 上述步骤发送控制指令给第二终端关闭副麦克风， 断开蓝牙语音链路。 另一种实施方式， 通话时， 第一终端蓝牙通讯装置与第二终端蓝牙通讯装置一直 建立语音链路连接， 第二终端副麦克风一直采集环境噪声发送给第一终端， 第一终端 检测通话环境确定是否进行双麦克风消噪处理。 另外， 第一终端的麦克风在通话时可作为副麦克风， 用于采集通话环境噪声并发 送给第二终端。 第二终端的麦克风作为主麦克风， 用于通话时采集语音及环境噪声， 第二终端对第一终端处理器传送过来的音频信号和其自身的麦克风采集到音频信号进 行 DSP消噪处理， 由于此两麦克风分离， 距离远， 采集的语音信号幅度差异大， 经处 理器 DSP消噪处理后可以保留良好的语音信号，此实施方式可有效解决现有双麦克风 消噪手机由于双麦克风距离有限导致免提通话时发送响度受嘴与手机距离影响问题。
当第一终端没有无线通信功能时， 本发明所述的通话降噪的方法与上述在原理上 基本相同， 步骤与上述第一终端通话降噪的方法基本相同， 不同之处在于： 在步骤 S103中，编码输出降噪后的通话语音至与第二终端，然后第二终端再将降 噪后的通话语音发送至与第一终端通信的设备； 在步骤 S104中，所述第一终端直接编码输出其采集的通话语音至第二终端，然后 第二终端再将降噪后的通话语音发送至与第一终端通信的设备。 在本实施方式中， 第一终端和第二终端都可以作为对第一终端采集的通话语音和 环境噪声进行降噪处理的处理方， 当第一终端作为降噪处理方时， 第一终端采集语音 通话和环境噪声直接发送给第一终端的处理器， 第一终端的处理器接收第二终端采集 的环境噪声， 然后第一终端根据第二终端发送过来的环境噪声对通话语音进行降噪处 理， 编码输出降噪后的通话语音至与第二终端， 然后第二终端再将降噪后的通话语音 发送至与第一终端通信的设备； 当第二终端作为降噪处理方时， 第一终端采集语音通 话和环境噪声发送给第二终端的处理器， 然后第二终端根据第一终端发送过来的通话 语音和环境噪声对通话语音进行降噪处理， 第二终端将降噪后的通话语音发送至与第 一终端通信的设备。 以上所述仅为本发明的优选实施例， 并非因此限制本发明的专利范围， 凡是利用 本发明说明书及附图内容所作的等效结构或流程变换， 或直接或间接运用在其它相关 的技术领域， 均同理包括在本发明的专利保护范围内。 工业实用性 本发明上述实施例提供的技术方案可以应用于通信领域的智能终端的消噪处理， 特别是适用于穿戴智能终端的消噪处理， 应用本发明的上述实施例提供的技术方案的 智能终端能有效的消除噪声， 提高通话质量。
Claims
权 利 要 求 书 、 一种智能终端通话降噪的方法， 包括： 在语音通信开始时， 第一终端采集通话语音和环境噪声；
所述第一终端根据其采集的通话语音或 /和环境噪声，判断是否处于安静状 态； 若所述第一终端不处于安静状态， 则所述第一终端根据第二终端发送过来 的环境噪声对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不做降噪处理。 、 如权利要求 1所述的方法， 其中， 所述第一终端根据其采集的通话语音或 /和环 境噪声， 判断是否处于安静状态的步骤包括：
根据所述第一终端采集的通话语音和环境噪声得出信噪比；
若所述信噪比大于等于预设值， 则所述第一终端处于安静状态。 、 如权利要求 1所述的方法， 其中， 所述第一终端根据其采集的通话语音或 /和环 境噪声， 判断是否处于安静状态的步骤包括：
根据所述第一终端采集的环境噪声得出环境噪声的幅度； 若所述环境噪声幅度值小于预设噪声信号幅度阀值， 则所述第一终端处于 安静状态。 、 如权利要求 1所述的方法， 其中， 所述若第一终端不处于安静状态， 则所述第 一终端根据所述第二终端发送过来的环境噪声进行降噪处理的步骤包括：
若所述第一终端不处于安静状态， 则建立与所述第二终端的语音链路， 启 动所述第二终端的麦克风进行采集环境噪声；
所述第一终端根据所述第二终端发送过来的环境噪声对通话语音进行降噪 处理。 、 如权利要求 1所述的方法，其中，所述第一终端不处于安静状态进行降噪处理， 或者所述第一终端处于安静状态不进行降噪处理的步骤之后还包括：
所述第一终端实时监测其所处环境是否是安静状态， 判断是否需要对通话 语音进行降噪；
若所述第一终端处于不安静状态， 则对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不对通话语音进行降噪处理。 、 如权利要求 1所述的方法， 其中， 在所述第一终端的主麦克风采集通话语音和 环境噪声的步骤之后， 还包括：
所述第一终端建立与所述第二终端的持久语音链路， 所述第二终端的麦克 风不断将采集到的环境噪声发送给所述第一终端。 、 一种智能终端， 包括： 声音采集单元， 设置为在语音通信开始时， 采集通话语音和环境噪声； 安静状态判断单元，设置为根据所述主麦克风采集的通话语音或 /和环境噪 声， 判断是否处于安静状态；
语音降噪输出单元， 设置为在所述第一终端不处于安静状态时， 根据所述 第二终端发送过来的环境噪声对通话语音进行降噪处理； 语音输出单元， 设置为在所述第一终端处于安静状态， 不对通话语音做降 噪处理。 、 如权利要求 7所述的智能终端， 其中， 所述安静状态判断单元设置为： 根据所述第一终端采集的通话语音和环境噪声得出信噪比；
若所述信噪比大于等于预设值， 则所述第一终端处于安静状态 、 如权利要求 7所述的智能终端， 其中， 所述安静状态判断单元设置为： 根据所述第一终端采集的环境噪声得出环境噪声的幅度； 若所述环境噪声幅度值小于预设噪声信号幅度阀值， 则所述第一终端处于 安静状态。 0、 如权利要求 7所述的智能终端， 其中， 所述语音降噪输出单元包括： 链路启用模块， 设置为在所述第一终端不处于安静状态时， 建立与所述第 二终端的语音链路， 启动所述第二终端的麦克风进行采集环境噪声；
降噪输出模块， 设置为根据所述第二终端发送过来的环境噪声对通话语音 进行降噪处理。 1、 如权利要求 7所述的智能终端， 其中， 所述智能终端还包括：
实时监测单元， 设置为实时监测所述第一终端所处环境是否是安静状态， 判断是否需要对通话语音进行降噪； 若所述第一终端处于不安静状态， 则对通话语音进行降噪处理； 若所述第一终端处于安静状态， 则不对通话语音进行降噪处理。 如权利要求 7所述的智能终端， 其中， 所述智能终端还包括： 持久链路建立单元， 设置为建立与所述第二终端的持久语音链路， 所述第 二终端的麦克风不断将采集到的环境噪声发送给所述第一终端。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410219218.6 | 2014-05-22 | ||
| CN201410219218.6A CN104092801A (zh) | 2014-05-22 | 2014-05-22 | 智能终端通话降噪方法及智能终端 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2015176372A1 true WO2015176372A1 (zh) | 2015-11-26 |
Family
ID=51640481
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2014/082285 Ceased WO2015176372A1 (zh) | 2014-05-22 | 2014-07-15 | 智能终端通话降噪方法及智能终端 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN104092801A (zh) |
| WO (1) | WO2015176372A1 (zh) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105516850A (zh) * | 2015-12-15 | 2016-04-20 | 上海斐讯数据通信技术有限公司 | 降噪耳机及降噪方法 |
Families Citing this family (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104318919B (zh) * | 2014-10-22 | 2020-04-21 | 上海斐讯数据通信技术有限公司 | 环境消噪方法、系统及移动终端 |
| CN104575509A (zh) * | 2014-12-29 | 2015-04-29 | 乐视致新电子科技（天津）有限公司 | 语音增强处理方法及装置 |
| CN106534510B (zh) * | 2016-11-02 | 2020-11-13 | 惠州Tcl移动通信有限公司 | 一种基于移动终端的麦克风通话控制方法及移动终端 |
| CN109417581A (zh) * | 2017-03-14 | 2019-03-01 | 华为技术有限公司 | 一种终端通话方法及终端 |
| CN106961509B (zh) * | 2017-04-25 | 2019-12-31 | Oppo广东移动通信有限公司 | 通话参数处理方法、装置及电子设备 |
| CN107644634A (zh) * | 2017-08-01 | 2018-01-30 | 柴世军 | 环境音频降噪装置 |
| CN107370883A (zh) * | 2017-08-04 | 2017-11-21 | 广东欧珀移动通信有限公司 | 改善通话效果的方法、装置及移动终端 |
| CN107623783A (zh) * | 2017-09-11 | 2018-01-23 | 安徽省未来博学信息技术有限公司 | 一种通话语音优化系统 |
| CN108063871A (zh) * | 2017-12-19 | 2018-05-22 | 厦门美图移动科技有限公司 | 通话处理方法及装置 |
| CN109119091A (zh) * | 2018-08-17 | 2019-01-01 | 西安蜂语信息科技有限公司 | 语音通话降噪方法及装置 |
| CN109120790B (zh) * | 2018-08-30 | 2021-01-15 | Oppo广东移动通信有限公司 | 通话控制方法、装置、存储介质及穿戴式设备 |
| CN110602327B (zh) * | 2019-09-24 | 2021-06-25 | 腾讯科技（深圳）有限公司 | 语音通话方法、装置、电子设备及计算机可读存储介质 |
| CN110839108A (zh) * | 2019-11-06 | 2020-02-25 | 维沃移动通信有限公司 | 一种降噪方法及电子设备 |
| CN115022767A (zh) * | 2022-07-28 | 2022-09-06 | 歌尔科技有限公司 | 耳机降风噪方法、装置、耳机及计算机可读存储介质 |
| CN118262735B (zh) * | 2024-05-28 | 2024-08-09 | 广东鼎创智造科技有限公司 | 一种麦克风自动降噪系统 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101699831A (zh) * | 2009-10-23 | 2010-04-28 | 深圳华为通信技术有限公司 | 终端送话方法、系统及设备 |
| CN102209135A (zh) * | 2011-05-20 | 2011-10-05 | 中兴通讯股份有限公司 | 用于对移动终端降风噪的装置及方法 |
| CN103325378A (zh) * | 2013-07-16 | 2013-09-25 | 珠海市魅族科技有限公司 | 一种终端语音降噪方法及终端 |
| CN103716438A (zh) * | 2012-09-28 | 2014-04-09 | 联想移动通信科技有限公司 | 降噪方法，装置及移动终端 |
| CN103811013A (zh) * | 2012-11-07 | 2014-05-21 | 中国移动通信集团公司 | 噪声抑制方法、装置、电子设备和通信处理方法 |
-
2014
- 2014-05-22 CN CN201410219218.6A patent/CN104092801A/zh active Pending
- 2014-07-15 WO PCT/CN2014/082285 patent/WO2015176372A1/zh not_active Ceased
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101699831A (zh) * | 2009-10-23 | 2010-04-28 | 深圳华为通信技术有限公司 | 终端送话方法、系统及设备 |
| CN102209135A (zh) * | 2011-05-20 | 2011-10-05 | 中兴通讯股份有限公司 | 用于对移动终端降风噪的装置及方法 |
| CN103716438A (zh) * | 2012-09-28 | 2014-04-09 | 联想移动通信科技有限公司 | 降噪方法，装置及移动终端 |
| CN103811013A (zh) * | 2012-11-07 | 2014-05-21 | 中国移动通信集团公司 | 噪声抑制方法、装置、电子设备和通信处理方法 |
| CN103325378A (zh) * | 2013-07-16 | 2013-09-25 | 珠海市魅族科技有限公司 | 一种终端语音降噪方法及终端 |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105516850A (zh) * | 2015-12-15 | 2016-04-20 | 上海斐讯数据通信技术有限公司 | 降噪耳机及降噪方法 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104092801A (zh) | 2014-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2015176372A1 (zh) | 智能终端通话降噪方法及智能终端 | |
| US9756422B2 (en) | Noise estimation in a mobile device using an external acoustic microphone signal | |
| DK2882204T3 (en) | Hearing aid device for hands-free communication | |
| CN111464905A (zh) | 基于智能穿戴设备的听力增强方法、系统和穿戴设备 | |
| JP6450458B2 (ja) | 自身の声を迅速に検出する方法と装置 | |
| CN202889458U (zh) | 一种根据环境噪声自动调节通话音量的手机 | |
| US20150016633A1 (en) | Method and apparatus for multi-stage adaptive volume control | |
| CN108886647B (zh) | 耳机降噪方法及装置、主耳机、从耳机及耳机降噪系统 | |
| US20150110263A1 (en) | Headset Dictation Mode | |
| WO2013155777A1 (zh) | 一种无线会议终端及其进行语音信号传递的方法 | |
| CN108551604B (zh) | 一种降噪方法、降噪装置及降噪耳机 | |
| CN112992169A (zh) | 语音信号的采集方法、装置、电子设备以及存储介质 | |
| CN110896509A (zh) | 耳机佩戴状态确定方法、电子设备控制方法及电子设备 | |
| WO2014117722A1 (zh) | 语音处理方法、装置及终端设备 | |
| CN113542960B (zh) | 音频信号处理方法、系统、装置、电子设备和存储介质 | |
| US10631108B2 (en) | Hearing augmentation systems and methods | |
| US20250024210A1 (en) | Binaural hearing aid system and a hearing aid comprising own voice estimation | |
| TR201807595T4 (tr) | Bir iletişim sistemi içinde ses sinyali işleme. | |
| CN105430546A (zh) | 耳机及其控制方法 | |
| CN106448691A (zh) | 一种用于扩音通信系统的语音增强方法 | |
| WO2015058484A1 (zh) | 降噪耳机及其降噪方法 | |
| WO2015180249A1 (zh) | 音频信号的消噪方法及系统 | |
| CN109218879A (zh) | 头戴式耳机、用于头戴式耳机的方法以及计算机可读介质 | |
| CN112822585A (zh) | 一种入耳式耳机的音频播放方法、装置及系统 | |
| WO2018005140A1 (en) | Hearing augmentation systems and methods |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 14892438 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 14892438 Country of ref document: EP Kind code of ref document: A1 |