JP7179128B1 - Wireless communication device and wireless communication system - Google Patents

Abstract

A voice period can be detected with high accuracy and wireless transmission can be started. A wireless communication device includes a voice signal detector that compares an input voice signal with a threshold value and outputs a voice detection signal that is a comparison result signal, and modulates a radio carrier according to the input voice signal. and a control unit for controlling the modulation transmission unit to be in a transmission state based on the voice detection signal. A wireless communication apparatus includes a noise canceller that performs audio signal processing to cancel noise from an input audio signal and outputs the processed audio signal to a modulation transmitter. The audio signal detection section is provided after the noise cancellation section, compares the audio signal from the noise cancellation section with a threshold value, and outputs an audio detection signal. The noise cancellation unit performs noise cancellation processing using a deep learning model unit that is learned using characteristic parameters of human speech and determines whether or not it is a non-speech period containing noise from the demodulated speech signal. . [Selection drawing] Fig. 1

Description

The present invention relates to a wireless communication device having, for example, a VOX (Voice Operated Transmitter) circuit and a wireless communication system including a plurality of wireless communication devices.

A VOX circuit is provided in a wireless communication device according to the prior art. Here, the VOX circuit is a circuit that controls ON/OFF of the output of the transmission signal according to the presence or absence of voice. is not transmitted, and the wireless communication device is controlled to be in a reception state.

For example, Patent Literature 1 proposes a voice segment detection device having the following configuration in order to detect voice segments with high accuracy and improve call quality. In this voice interval detection device, the frequency distribution calculator calculates the frequency distribution of the input signal, and the flatness calculator calculates the flatness of the frequency distribution from the frequency distribution. For example, the average of the frequency distribution is obtained, and the sum of the differences between the frequency distribution and the average is used as the flatness of the frequency distribution. Furthermore, the voice/noise determination unit compares the flatness of the frequency distribution with a threshold value to determine whether the signal is voice or noise, and detects the voice section of the input signal.

Japanese Patent No. 3963850

7A and 7B are block diagrams showing an operation example of the wireless device 100 having the control section 110 according to the conventional example. Here, the audio signal detector 17 constitutes a so-called VOX circuit. As shown in FIGS. 7A and 7B, the audio signal detector 17, which is a VOX circuit mounted in the conventional wireless device 100, amplifies, filters, rectifies, and converts the audio signal from the microphone 12, for example. The audio signal level is compared with a predetermined threshold and radio transmission is initiated when the audio signal level is equal to or greater than the threshold. In the case of this conventional example, there have been many cases in which erroneous transmission is started due to erroneous recognition of an audio signal from the original microphone 12 due to wind noise, ambient noise, etc. (FIG. 7B). Therefore, for the above reasons, it has been said that the conventional VOX circuit cannot be used outdoors or in a noisy field.

That is, even in a wireless communication device having a conventional voice interval detection device, wind noise, ambient noise, and the like frequently cause erroneous detection of voice from the original microphone and start wireless transmission. There is a problem that the detection accuracy of the interval is still low.

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-described problems, and to provide a wireless communication device capable of detecting a voice period and starting wireless transmission with higher accuracy than the conventional example, and wireless communication including a plurality of wireless communication devices. It is to provide a system.

A wireless communication device according to an aspect of the present invention includes:
a first audio signal detector that compares an input audio signal with a predetermined first threshold value and outputs a first audio detection signal that is a comparison result signal;
a modulation transmitter that modulates a radio carrier wave according to the input audio signal and radio-transmits the radio signal;
A wireless communication device comprising: a control unit that controls the modulation transmission unit to be in a transmission state based on the first voice detection signal,
A noise canceling unit that performs audio signal processing so as to cancel noise from the input audio signal and outputs it to the modulation transmission unit;
The first audio signal detection unit is provided after the noise canceling unit, compares the audio signal from the noise canceling unit with the first threshold value, and performs first audio detection as a comparison result signal. output a signal,
The noise cancellation unit learns using a feature parameter of human speech, and uses a deep learning model unit that determines whether or not it is a non-speech period containing noise from the input speech signal, and performs noise cancellation processing. and
The wireless communication device
further comprising a second audio signal detection unit that compares an input audio signal with a predetermined second threshold value and outputs a second audio detection signal that is a comparison result signal;
The controller controls the modulation transmitter to start radio transmission based on the second voice detection signal input prior to the first voice detection signal.

Therefore, according to the wireless communication device of the present invention, the noise canceling unit performs audio signal processing so as to cancel noise from the input audio signal and outputs the audio signal to the modulation transmitting unit, A signal detection section is provided after the noise cancellation section, compares the audio signal from the noise cancellation section with the first threshold value, and outputs a first audio detection signal that is a comparison result signal. configured to As a result, it is possible to detect the voice period and start radio transmission with higher accuracy than in the conventional example.

1 is a block diagram showing a configuration example of a wireless device 1 according to Embodiment 1; FIG. 2 is a block diagram showing a configuration example of a noise cancellation unit 14 of FIG. 1; FIG. 3 is a block diagram showing a configuration example of a deep learning model unit 35 of FIG. 2; FIG. FIG. 10 is a block diagram showing a configuration example of a wireless device 1A according to Embodiment 2; FIG. 5 is a flow chart showing a first part of VOX mode transmission control processing executed by the control unit 10A of FIG. 4; FIG. FIG. 5 is a flow chart showing a second part of the VOX mode transmission control process executed by the control unit 10A of FIG. 4; FIG. 2 is a timing chart of each signal showing an operation example of the wireless device 1 of FIG. 1; 5 is a timing chart of each signal showing an operation example 1 of the wireless device 1A of FIG. 4; 5 is a timing chart of each signal showing an operation example 2 of the wireless device 1A of FIG. 4; FIG. 11 is a block diagram showing an operation example 1 of the wireless device 100 according to the conventional example; FIG. 11 is a block diagram showing an operation example 2 of the radio device 100 according to the conventional example; 2 is a block diagram showing an operation example 1 of the wireless device 1 of FIG. 1; FIG. 2 is a block diagram showing an operation example 2 of the wireless device 1 of FIG. 1; FIG. 3 is a block diagram showing an operation example 3 of the wireless device 1 of FIG. 1; FIG. 5 is a block diagram showing an operation example 1 of the wireless device 1A of FIG. 4; FIG.

Hereinafter, embodiments and modifications according to the present invention will be described with reference to the drawings. In addition, the same code|symbol is attached|subjected about the same or similar component.

(Embodiment 1)
FIG. 1 is a block diagram showing a configuration example of a wireless device 1 according to an embodiment. In FIG. 1, a wireless device 1 is an example of a wireless communication device, and includes a receiving antenna 21, a receiving demodulator 22, an audio signal amplifier 23, a speaker 24, a control unit 10, and a PTT (Push To Talk) key. 11A and VOX mode key 11B; a microphone 12; an audio signal amplifier 13; a noise canceling unit 14; a modulation transmission unit 15; be done. Here, the PTT key 11A is turned on when it is desired to transmit voice, and the VOX mode key 11B is a VOX mode using the voice signal detector 17, which is an example of a VOX circuit (for example, a wireless signal is transmitted only when voice is detected). However, when there is no sound around the microphone, no radio signal is transmitted, and the radio communication apparatus is controlled to be in a receiving state). 15 and the operation of the reception demodulator 22.

Here, the wireless device 1 according to the embodiment is an example of a wireless communication device of a specified low-power wireless station for a specified low-power wireless communication system, for example. In this embodiment, the wireless device 1 performs noise cancellation processing using the deep learning model unit 35 (FIG. 3) at the front stage of the audio signal detection unit 17, which is a VOX circuit, in the transmission unit, for example, FM (frequency modulation) or PM (phase modulation), which is particularly effective in reducing noise. Also, a plurality of wireless devices 1 constitute a wireless communication system.

In FIG. 1, a radio signal received by a receiving antenna 21 is input to a receiving demodulator 22 . After performing low-noise amplification, low-frequency conversion, intermediate frequency amplification, etc. on the received radio signal, the reception demodulation unit 22 demodulates the received radio signal by a predetermined demodulation method such as FM (frequency modulation) or PM (phase modulation). The signal is demodulated and output to the speaker 24 via the audio signal amplifier 23 .

The microphone 12 converts an input voice into a voice signal and outputs the voice signal to the noise cancellation section 14 via the voice signal amplifier 13 . The noise canceling unit 14 uses a deep learning model unit 35 (FIG. 3) that has been deep-learned using human speech, and allows the input audio signal to pass through only during the audio signal period, thereby canceling noise. After performing audio signal processing, it outputs to the modulation transmission section 15 and the audio signal detection section 17 . The audio signal detector 17 is an example of a VOX circuit including a comparator. A voice detection signal (comparison result signal) S1 is output to the control unit 10, and an L level voice detection signal (comparison result signal) S1 is sent to the control unit 10 when the input voice signal is less than the threshold value. Output.

The control unit 10 is composed of a processor such as a CPU (Central Processing Unit) or DSP (Digital Signal Processor). After modulating the radio carrier wave according to the predetermined modulation method according to the received voice signal, the radio signal, which is the modulated radio carrier wave, is transmitted from the transmission antenna 16 after high frequency conversion and power amplification. In addition, the control unit 10 controls the operations of the modulation transmission unit 15 and the reception demodulation unit 22 in the VOX mode, that is, in response to the H-level audio detection signal S1 from the audio signal detection unit 17, the control unit 10 controls the modulation transmission unit 15. It is turned on to enter a transmission state, and in response to the L-level voice detection signal S1, the modulation transmission section 15 is turned off to enter a transmission stop state.

In this embodiment, the wireless device 1 operates in a simultaneous call system in which the transmission frequency and the reception frequency are different, but the present invention is not limited to this, and the wireless device 1 uses the same transmission frequency and reception frequency. frequency is used, the control unit 10 stops the operation of the reception demodulation unit 22 when the PTT key 11A is turned on or in response to the H level voice detection signal S1.

Next, with reference to FIG. 2, the configuration and operation of the noise cancellation unit 14 of FIG. 1 using the deep learning model unit 35 will be described below.

FIG. 2 is a block diagram showing a configuration example of the noise cancellation section 14 of FIG.

Here, the term "phoneme" refers to the unit of sound that distinguishes one word from another in a particular language, and the term "vibration rate" refers to the zero point of the digitized vibration data at each second. and 1, and the term "vibration count (VC)" refers to the sum of the values of the digitized vibration data within each frame. Further, the "vibration pattern" means the data distribution of the sum of the vibration frequencies calculated for each predetermined number of frames along the time axis. In the deep learning model unit 35, noise cancellation processing is performed in consideration of different vibration patterns, that is, differences in the data distribution of the sum of different vibration count values (VS values), and the vibration rate is similar to the vibration count value. However, the higher the vibration rate, the higher the vibration count.

Both the amplitude and vibration rate of the speech signal are observable. A feature of the noise cancellation unit 14 is to detect audio events according to the amplitude and vibration rate of the audio signal. Another feature is to distinguish between speech and non-speech/silence by measuring the sum of the vibration count values of the digitized vibration data for a predefined number of frames. Another feature is to classify the input audio signal data stream into different phonemes according to their vibration patterns. Another feature is the correct discrimination of the first activation phoneme from the incoming audio signal data stream to trigger downstream processing units, thereby calculating power consumption, etc. of the computing system containing the processing units. It is to save costs.

In FIG. 2, the noise cancellation unit 14 performs noise cancellation processing using audio event detection, and includes an audio signal preprocessing unit 38, an AD converter 39, and an audio signal processing unit 30. be done. Here, the audio signal preprocessing unit 38 performs audio signal preprocessing, including high-pass filtering, low-pass filtering, amplification, or a combination thereof, on the analog audio signal, and outputs the processed analog audio signal. Output to AD converter 39 . That is, the audio signal pre-processing unit 38 allows the audio signal from the microphone 12 to pass only within a predetermined level range of human audio signals and in a predetermined bandwidth. Next, the AD converter 39 AD-converts the analog audio signal into a digital audio signal according to a predetermined reference voltage Vref and an allowable voltage Vadm (<Vref), and outputs the digital audio signal to the input interface 36 of the audio signal processing section 30 .

In the present embodiment, in the AD converter 39, the allowable voltage Vadm smaller than the reference voltage Vref is combined with the reference voltage Vref to obtain the first threshold voltage Vth1 (=Vref+Vadm)) and the second threshold voltage Vth2 (=Vref−Vadm), and the AD converter 39 converts the first threshold voltage Vth1 based on the first threshold voltage Vth1 and the second threshold voltage Vth2. Noise and interference in the input analog audio signal are removed by not performing AD conversion on noise above or below the second threshold voltage Vth2, but by performing AD conversion on the audio signal between them. can do. Here, if Vref=1.0 V and Vadm=0.01 V, for example, it can be understood that the frequency of vibration data is low in a quiet environment, and the frequency of vibration data is high in a voice environment. In this embodiment, the “frame size” means the number of sampling points corresponding to the digitized vibration data in each frame, and the “phoneme window Tw” means the speech feature amount of each phoneme. Means time to collect. In a preferred embodiment, the duration Tf of each frame is eg 0.1-1 milliseconds (ms) and the phoneme window Tw is eg about 0.3 seconds. In a further preferred embodiment, the number of sampling points corresponding to the digitized vibration data within each frame ranges, for example, from 1-16.

When analyzing speech signals, the method of short-term analysis is usually adopted since most speech signals are stable in a short period of time. For example, assuming that the sampling frequency fs used by the AD converter 39 is 16000 and the duration Tf of each frame is 1 ms, the frame size is fs×1/1000=16 sample points.

In FIG. 2, the audio signal processing unit 30 is configured by, for example, a computer device,
(1) A CPU (Central Processing Unit) 31 that executes predetermined audio signal processing such as noise cancellation;
(2) a ROM (Read Only Memory) 32 that stores an operating system that executes the basic processing of the CPU 31, the audio signal processing program, and data necessary for executing the program;
(3) a RAM (Read Access Memory) 33 for storing data being processed when the operating system for executing the basic processing of the CPU 31 and the program for the audio signal processing are executed;
(4) a non-volatile EEPROM (Electrically Erasable Programmable Memory) 34 for storing later-described setting data necessary for executing the audio signal processing;
(5) For example, it is composed of a neural network, etc., and removes noise from voice signal data that is input after deep learning based on human voice signal data, and extracts and outputs substantially only the voice signal. a deep learning model unit 35;
(6) an input interface 36 that performs a predetermined signal conversion process for converting audio signal data input from the AD converter 39 into a signal specification value in the subsequent stage and outputs the result to the CPU 31;
(7) The audio signal data from which noise has been removed by the deep learning model unit 35 is subjected to a predetermined signal conversion process for converting it into a signal specification value in the subsequent stage, and the wireless device 1 is transmitted through the terminal T12, the audio line L2, etc. an output interface 37 that outputs to
configured with

Here, the EEPROM 34 stores, for example, a series of vibration count values VC, a sum of vibration count values VS, a sum of vibration count values VSf, a sum of vibration count values VSp (to be described later), and sound feature values of all feature vectors. do. Note that the EEPROM 34 may be a storage device such as an external memory. The audio event detection method applied to the audio signal processor 30 is executed during runtime by the CPU 31 to capture audio events. Audio event detection is performed assuming fs=16000, Tf=1 ms, and Tw=0.3 s.

Specifically, the CPU 31 calculates the sum of the vibration data values of the current frame (that is, within 1 ms) to be processed, acquires the vibration count value VC, and then obtains the VC of the current frame at time Tj. Store the value in EEPROM 34 . Here, the vibration count values VC of x frames are added to obtain the sum of the vibration count values VS of the current frame at time Tj. x frames includes the current frame. In one embodiment, the CPU 31 adds the vibration count value VC of the current frame at the time Tj and the sum VSp of the vibration count values of the (x−1) frames immediately before that frame to obtain the x number of vibration count values at the time Tj. Obtain the sum VS (=VC+VSp) of the vibration count values of the frame.

In the modified example, the CPU 31 outputs the vibration count value VC of the current frame at the time Tj, the total sum VSf of the vibration count values of the y frames immediately after that, and the (xy-1) frames immediately before that. to obtain the sum of vibration counts VS (=VC+VSf+VSp) of x frames at time Tj, where y is greater than or equal to zero. The CPU 31 stores the values of VS, VSf and VSp in the EEPROM 34 . In the preferred embodiment, the duration of x frames (phoneme window Tw) (x×Tf) is approximately 0.3 seconds. In a further preferred embodiment, the number of sampling points corresponding to x frames of digitized vibration data is in the range of x to 16x.

In general, with respect to speech signal data, the vibration patterns of the vibration count values VC are similar for the same phoneme, but the vibration patterns of the VS values are completely different for different phonemes. Therefore, the vibration pattern of the vibration count value VC can be used to distinguish between phonemes. In particular, for example, the barking of a chicken or a cat and human speech are completely different in terms of the frequency distribution of the vibration coefficient VC, and it is known that most of the vibration coefficient VC of human speech is distributed below 40. .

In the learning phase, the CPU 31 of the speech signal processing unit 30 first executes a predetermined speech signal data collection method a plurality of times to collect a plurality of feature vectors corresponding to a plurality of phonemes, and labels corresponding to the plurality of feature vectors. Append to form multiple labeled training examples. A plurality of labeled training examples for different phonemes, including the starting phoneme, are then applied to training the deep learning model unit 35 . Finally, a trained deep learning model unit 35 (constituting a predictive model for the speech signal data) is created to classify whether the incoming speech signal data stream contains the activation phoneme. When a predetermined phoneme is specified as a starting phoneme for the audio signal processing unit 30, the deep learning model unit 35 is trained with a plurality of labeled training examples for different phonemes including at least the specified phoneme.

That is, in the training phase, the deep learning model portion 35 is trained using a set of labeled training examples, whereby the deep learning model portion 35 learns the three audio features of each frame of the labeled training examples. Based on the quantity (eg, (VSj, TDj, TGj)), we try to recognize a given activation phoneme between j=0-299. At the end of the learning phase, the trained deep learning model unit 35 provides a learned score corresponding to the activation phoneme, which is then converted at runtime to the incoming audio signal data stream. Used as criteria for classification. VSj, TDj, and TGj are defined as follows.
(1) VSj: sum of vibration count values (VS value) of frame j;
(2) TDj: the time period of the sum of non-zero vibration counts (VS values) at frame j; and (3) TGj: the time between the sums of non-zero vibration counts (VS values) at frame j. Gap (time gap).

Various machine learning techniques related to supervised learning can be used to train the deep learning model portion 35, such as support vector machine (SVM) methods, random forest methods, convolutional neural network methods, etc. can. In supervised learning, a function calculator (i.e., deep learning model portion 35) is created using a plurality of labeled training examples, each of which consists of an input feature vector and a labeled output. When trained, deep learning model portion 35 can be applied to new unlabeled examples to generate corresponding scores or predictions.

FIG. 3 is a block diagram showing a detailed configuration example of the deep learning model unit 35 of FIG.

The deep learning model unit 35 is implemented using a neural network, as shown in FIG. 3, for example. Here, the neural network comprises one input layer 41 , at least one and preferably several intermediate layers 42 and one output layer 43 . The input layer 41 has three input neurons 51, 52, 53, each input neuron 51, 52, 53 corresponding to three audio feature values (i.e., VSj, TDj, TGj) for each frame of the feature vector. . The intermediate layer 42 is also composed of neurons 61-74 having weight coefficients associated with each input neuron 51, 52, 53 and bias coefficients for each neuron. By changing the weighting and biasing factors of each neuron 61-74 of the hidden layer 42 through cycles of the learning phase, the neural network can be trained to report a predicted value for a given type of input. . In addition, the output layer 43 includes one output neuron 81 that provides one prediction value corresponding to a phoneme (specifically indicating whether it is a speech period or a non-speech period containing noise).

As described above, according to the present embodiment, in the noise cancellation unit 14, the deep learning model unit 35 is learned using the feature parameters of human speech, and the non-speech period including noise is obtained from the input speech signal. It is determined whether or not. Then, the CPU 31 of the audio signal processing unit 30 performs noise cancellation processing so as not to pass the non-speech period containing noise from the input audio signal based on the determination of the deep learning model unit 35, and performs the noise cancellation. Outputs the processed audio signal. Here, the deep learning model unit 35 receives as input the characteristic parameters of human speech, and outputs the determination result of determining whether or not the input speech signal is a non-speech period containing noise. It consists of a neural network.

8A and 8B are block diagrams showing operation examples 1 and 2 of the wireless device 1 of FIG. 1, respectively.

In the present embodiment, the wireless device 1 includes the noise cancellation section 14 in the transmitting section in front of the audio signal detection section 17, which is a VOX circuit. By doing so, it becomes a transmission state (FIG. 8A). On the other hand, with wind noise or noise, the audio signal detector 17 does not detect an audio signal above a certain level, so the transmission state is not established (FIG. 8B). Therefore, according to the present embodiment, the noise canceling unit 14 greatly improves the detection accuracy of the speech period, and the sound from the microphone 12 is not mistaken for the original speech due to wind noise, ambient noise, etc., as in the conventional example. It is possible to effectively prevent the detection and the start of wireless transmission, and it is possible to detect the voice period and start the wireless transmission with higher accuracy than in the conventional example.

Furthermore, in the present embodiment, the radio 1 includes a noise cancellation section 14 in its transmission section, which is particularly effective in reducing noise in, for example, FM (Frequency Modulation) or PM (Phase Modulation). As a result, in a wireless communication device that performs wireless communication, noise cancellation can be performed so as to effectively output a human uttered voice, compared to the conventional technology. By providing the noise canceling unit 14 on the transmission side, noise can be effectively removed from the audio signal in circuits and devices (for example, wireless relay devices) on and after the transmission side.

Next, problems in the first embodiment will be explained below.

6A is a timing chart of each signal showing an operation example of the wireless device 1 according to Embodiment 1, and FIG. 9 is a block diagram showing an operation example 3 of the wireless device 1. FIG.

Since the noise canceling unit 14 performs processing such as A/D conversion, D/A conversion, memory writing/reading, etc., the output signal is always delayed with respect to the input signal, and the voice period detection time T1 occurs. Therefore, as shown in FIG. 6A, the audio signal output from the noise canceling unit 14 is delayed due to the audio period detection time T1 (t1 to t2) of the noise canceling unit 14, and the level of the audio signal detecting unit 17 Due to the detection time T2 (generally <T1 (because the operation time of the comparator is shorter than the operation time of the CPU, etc.); t2 to t3), the audio signal output from the audio signal detector 17 is delayed. At this time, there is a possibility that the beginning of the voice signal will be truncated for at least time T2 in the receiving section of another receiver that receives the signal transmitted from the modulation transmitting section 15 . A second embodiment for solving this problem will be described below.

In FIG. 6A, due to the voice period detection time T1a (≈T1; t4 to t5) of the noise canceling unit 14, the voice signal from the noise canceling unit 14 will continue to be output from time t4 when the voice signal to the microphone 12 disappears. The output end is delayed.

(Embodiment 2)
FIG. 4 is a block diagram showing a configuration example of the wireless device 1A according to the second embodiment. 4, the wireless device 1A according to the second embodiment has the following differences compared to the wireless device 1 of FIG. 1 according to the first embodiment.
(1) An audio signal detection section 18 is further provided in the stage after the audio signal amplifier 13 and in the stage before the noise cancellation section 14 .
(2) Instead of the control unit 10, a control unit 10A for executing the VOX mode transmission control process shown in FIGS. 5A and 5B is provided. In the analysis of the noise canceling unit 14 and the audio signal detecting unit 17, the control unit 10A determines that "there is no audio signal" after receiving the wireless signal for a predetermined time T12 (t22 to t23 in FIG. 6C; T12 (predetermined at the circuit design stage).
Differences will be described below.

4, the audio signal detector 18 includes a comparator, and the audio signal from the audio signal amplifier 13 is detected by a predetermined threshold (even if the threshold is the same as the threshold of the audio signal detector 17). Alternatively, for example, the threshold value of the audio signal detection unit 18 may be set to be smaller or larger than the threshold value of the audio signal detection unit 17, and may be set differently). is equal to or greater than the threshold value, an H-level voice detection signal (comparison result signal) S2 is output to the control unit 10A, while an L-level voice signal is output to the control unit 10A when the input voice signal is less than the threshold value. A detection signal (comparison result signal) S2 is output to the control section 10A.

The control unit 10A operates the modulation transmission unit 15 when the PTT key 11A is turned on, and the modulation transmission unit 15 modulates the radio carrier wave according to the predetermined modulation method in accordance with the input audio signal. A radio signal, which is a radio carrier wave, is transmitted from the transmission antenna 16 after high-frequency conversion and power amplification. Further, the control section 10A executes the transmission control processing shown in FIGS. 5A and 5B when controlling the operations of the modulation transmission section 15 and the reception demodulation section 22 in the VOX mode.

5A and 5B are flow charts showing VOX mode transmission control processing executed by the control section 10A of FIG. The detection flag F2 is a flag indicating whether the voice detection signal S2 has become H level or L level.

In the initialization process of step S1 in FIG. 5A, transmission is being stopped, the detection flag F2 is set to L level, the voice detection signals S1 and S2 are at L level, and the timer is reset. In step S2, it is determined whether or not the voice detection signal S2 has been received from the voice signal detector 18. If YES, the process proceeds to step S5, and if NO, the process proceeds to step S3. In step S3, the detection flag F2 is set to L level, and after resetting the timer in step S4, the process returns to step S2.

In step S5, it is determined whether or not the detection flag F2 is at L level. If YES, the process proceeds to step S6, and if NO, the process proceeds to step S11 in FIG. 5B. In step S6, the detection flag F2 is set to H level, in step S7 the control section 10A puts the modulation transmission section 15 into the transmission state to start transmission, and in step S8 time T12 is set in the timer to start counting. and proceed to step S11 in FIG. 5B.

In step S11 of FIG. 5B, it is determined whether or not the voice detection signal S1 has been received from the voice signal detection unit 17. If YES, the process proceeds to step S12, and if NO, the process proceeds to step S15. At step S12, after the timer is forcibly set to the state of expiration of the count, at step S13, it is determined whether or not the modulation transmission unit 15 is transmitting. If YES, the process returns to step S11. proceed to In step S14, the control section 10A starts radio transmission by setting the modulation transmission section 15 to the transmission state, and returns to step S11.

In step S15, it is determined whether or not the count of the timer has expired. If YES, the process proceeds to step S16, and if NO, the process returns to step S2 in FIG. 5A. Next, in step S16, it is determined whether or not the modulation transmitter 15 is transmitting. If YES, the process proceeds to step S17. If NO, the process returns to step S2 in FIG. 5A. In step S17, the control unit 10A stops transmission by putting the modulation transmission unit 15 into a transmission stop state, and then returns to step S2 in FIG. 5A.

FIG. 6B is a timing chart of each signal showing operation example 1 (when only human voice is input to the microphone 12) of the wireless device 1A of FIG. FIG. 6C is a timing chart of each signal showing operation example 2 (when noise or only noise is input to the microphone 12) of the wireless device 1A of FIG.

Embodiment 2 of FIG. 4 is characterized in that another audio signal detection section 18 is provided in the preceding stage of the noise cancellation section 14 . As shown in FIG. 6B, the signal processing time T1 (t1 to t3) of the noise canceling unit 14 composed of a CPU or the like is the delay from the time the audio signal detection unit 18 composed of a comparator detects the sound to the start of transmission. Sufficiently longer than time T11 (t1 to T11), the modulation transmitter 15 can start wireless transmission at time t11 before the voice detection signal S1 after noise cancellation is output (t3). As a result, the receiving unit of another receiver that receives the signal transmitted from the modulation transmitting unit 15 can output the "audio signal" as a received sound (101 in FIG. 6B).

In other words, in the second embodiment, wireless transmission is started in advance at the input terminal of the audio signal detection unit 18 in the preceding stage of the noise cancellation unit 14 regardless of the audio signal and noise. After that, the noise canceling analysis of the noise canceling unit 14 is completed, and an audio signal is output from the noise canceling unit 14 . Since the transmission has already been performed at this time, wireless transmission can be performed smoothly (FIG. 6B).

Also in FIG. 6C when noise or only noise is input to the microphone 12, the input terminal of the audio signal detection unit 18 preceding the noise canceling unit 14 starts wireless transmission in advance regardless of whether it is voice or noise. After that, even if the audio signal is not detected at the input end of the audio signal detector 17, it is determined that there is no audio signal and only noise, and the transmission is stopped. At that time, the signal is temporarily transmitted (T12), but since there is no modulated voice signal to be transmitted, the demodulated sound of the wireless device on the receiving side remains silent (T13 (=T12)).

FIG. 10 is a block diagram showing an operation example 1 of the wireless device 1A of FIG.

As described above, according to the second embodiment, since the separate audio signal detection unit 18 is provided before the noise canceling unit 14, the noise-cancelled audio detection signal S1 is output as shown in FIG. before the modulation transmitter 15 can start radio transmission. As a result, the receiving section of another receiver that receives the signal transmitted from the modulation transmitting section 15 can output the "audio signal" as a received sound smoothly.

As described in detail above, according to the wireless communication apparatus of the present invention, the wireless device 1 includes the noise canceling section 14 in the transmitting section before the audio signal detecting section 17, which is a VOX circuit. When the detection unit 17 detects a certain level or higher only from the sound, the transmission state is set (FIG. 8A). On the other hand, with wind noise or noise, the audio signal detector 17 does not detect an audio signal above a certain level, so the transmission state is not established (FIG. 8B). Therefore, the noise canceling unit 14 greatly improves the detection accuracy of the voice signal section, and radio transmission is performed by erroneously detecting the voice from the original microphone 12 due to wind noise, ambient noise, etc., as in the conventional example. It is possible to effectively prevent the start, and to detect the voice period with higher accuracy than the conventional example and start the wireless transmission.

Further, since another audio signal detection unit 18 is provided in the preceding stage of the noise canceling unit 14, as shown in FIG. 10, wireless transmission can be started before the noise-cancelled audio detection signal S1 is output. can. As a result, the receiving unit of another receiver that performs reception can output the "audio signal" as a received sound smoothly.

1, 1A Wireless device 10, 10A Control unit 11 Operation unit 11A PTT key 11B VOX mode key 12 Microphone 13 Audio signal amplifier 14 Noise cancellation unit 15 Modulation transmission unit 16 Transmission antennas 17, 18 Audio signal detection unit 21 Reception antenna 22 Reception demodulation Unit 23 Audio signal amplifier 24 Speaker 30 Audio signal processing unit 31 CPU
32 ROMs
33 RAM
34 EEPROMs
35 deep learning model unit 36 input interface 37 output interface 38 audio signal preprocessing unit 39 AD converter 41 input layer 42 intermediate layer 43 output layers 51 to 81 neuron 100 wireless device 110 control unit

Claims (7)

a first audio signal detector that compares an input audio signal with a predetermined first threshold value and outputs a first audio detection signal that is a comparison result signal;
a modulation transmitter that modulates a radio carrier wave according to the input audio signal and radio-transmits the radio signal;
A wireless communication device comprising: a control unit that controls the modulation transmission unit to be in a transmission state based on the first voice detection signal,
A noise canceling unit that performs audio signal processing so as to cancel noise from the input audio signal and outputs it to the modulation transmission unit;
The first audio signal detection unit is provided after the noise canceling unit, compares the audio signal from the noise canceling unit with the first threshold value, and performs first audio detection as a comparison result signal. output a signal,
The noise cancellation unit learns using a feature parameter of human speech, and uses a deep learning model unit that determines whether it is a non-speech period containing noise from the input speech signal, and performs noise cancellation processing. and
The wireless communication device
further comprising a second audio signal detection unit that compares an input audio signal with a predetermined second threshold value and outputs a second audio detection signal that is a comparison result signal;
The wireless communication device , wherein the controller controls the modulation transmitter to start wireless transmission based on the second voice detection signal input prior to the first voice detection signal . 2. The radio communication apparatus according to claim 1 , wherein said modulating transmission unit modulates a radio carrier wave according to a frequency modulation method or a phase modulation method according to an input audio signal. The noise cancellation unit, based on the determination of the deep learning model unit, performs noise cancellation processing so as not to pass a non-speech period containing noise from the input audio signal, and the audio signal after the noise cancellation processing. provided with an audio signal processing unit that outputs
The radio communication device according to claim 1 or 2 . The noise cancellation unit is
Further comprising an audio signal pre-processing unit provided in the preceding stage of the audio signal processing unit for passing only a predetermined bandwidth within a predetermined level range of a human audio signal to an input audio signal. 4. The wireless communication device of claim 3 . The deep learning model unit is composed of a predetermined neural network that receives characteristic parameters of human speech as an input and outputs a judgment result that determines whether or not the input speech signal is a non-speech period containing noise. The radio communication device according to any one of claims 1 to 4 , wherein The wireless communication device according to any one of claims 1 to 5 , wherein said wireless communication device is a specified low power radio station which is a wireless communication device for a specified low power wireless communication system. A wireless communication system comprising a plurality of wireless communication devices according to any one of claims 1-6 .

Images