JP6347029B2 - Intercom system - Google Patents

Description

  The present invention relates to an interphone system that performs a call by transmitting and receiving audio signals between an entrance cordless handset and a room base phone connected via any one of a wired transmission path and a wireless transmission path. In particular, the present invention relates to an intercom system having an audio signal recognition technology.

  Conventionally, as this type of intercom system, there has been disclosed a communication device that can receive an incoming call without being forcibly terminated even when an incoming call is received from an external line during an extension call or remote operation. According to this communication apparatus, it is possible to superimpose a ringing tone for notifying an incoming call in a call during an external line call or a signal during a remote operation (see, for example, Patent Document 1).

Japanese Patent Laid-Open No. 11-341292

  However, according to the communication device described in the background art, a ringing tone (ringing tone signal) can be superimposed on a voice signal for activating the intercom function, for example, as a signal during remote operation. No means for recognizing only the voice signal out of the voice signal on which (calling tone signal) is superimposed is provided.

  The present invention has been made in order to solve this problem, and when a ringing tone signal is output from a parent unit speaker of a living room parent unit, the ringing tone signal is superimposed and the parent unit microphone is superimposed. It is an object of the present invention to provide an intercom system that improves the recognition accuracy of a voice signal input to the TV.

In order to achieve the above-mentioned object, an interphone system according to a first aspect of the present invention includes an entrance cordless handset having a handset microphone and handset speaker, and any one of a wired transmission path and a wireless transmission path. There is provided a parent unit microphone having a parent unit microphone and a parent unit speaker for inputting / outputting a voice signal for performing a call between the child unit microphone and the child unit speaker of the entrance slave unit connected via a road. . The living room master unit analog / digital converts the transmission analog voice signal input to the base unit microphone into a transmission digital voice signal, and is input to the slave unit microphone of the entrance unit and transmitted via the transmission path. A master audio codec that digitally / analog converts the incoming digital audio signal into an incoming analog audio signal and outputs it from the parent speaker, and a transmitted digital audio signal sent from the parent audio codec to the transmission path and the parent from the transmission path And an audio processing unit that performs transmission / reception control on the received digital audio signal transmitted to the machine audio codec. The audio processing unit generates a test signal to be output from the main unit speaker when the power is turned on and sends the test signal to the main unit audio codec, and after the digital / analog conversion via the main unit audio codec, A recording circuit for storing a test signal transmitted after being acoustically fed back from the main unit speaker to the main unit microphone and analog / digital converted via the main unit audio codec, and a test signal and recording generated by the test signal generation circuit A transfer function calculation circuit for calculating an impulse response based on a test signal stored in the circuit, and a ringing tone signal generation circuit for generating a ringing tone signal to be output from the parent speaker at the time of calling and sending the ringing tone signal to the parent unit voice codec And the ringing tone signal generated by the ringing tone signal generating circuit is convoluted with the impulse response calculated by the transfer function calculating circuit. A suppression signal generation circuit that generates a suppression signal for suppressing the feedback data, and a delay circuit that transmits the suppression signal generated by the suppression signal generation circuit after being delayed by an acoustic feedback time from the parent speaker to the parent microphone. , Generated by the ringing tone signal generation circuit at the time of calling, and acoustically fed back from the parent speaker to the parent microphone via the parent unit voice codec, and then superimposed on the ringing tone signal transmitted via the parent unit voice codec A subtracting circuit that subtracts the suppression signal sent from the delay circuit from the voice signal for activating the interphone function, and an interphone function on which the ringing tone signal is superimposed based on the output signal from the subtracting circuit And a speech recognition processing circuit for recognizing the speech signal from the speech signal.

  According to the interphone system of this aspect, when the system is turned on, the test signal to be output from the master speaker of the living room master is generated by the test signal generation circuit of the voice processing unit, and the test signal is output from the master speaker. An impulse response with the test signal stored in the recording circuit after being acoustically fed back to the parent microphone can be calculated by the transfer function calculation circuit. In addition, when the system is called, a ringing tone signal to be output from the parent speaker is generated by the ringing tone signal generation circuit, and an arithmetic process for convolving the impulse response calculated by the transfer function calculation circuit with the ringing tone signal is suppressed. The delay circuit delays the suppression signal generated by the suppression signal generation circuit by an acoustic feedback time from the parent speaker to the parent microphone and sends it to the subtraction circuit. The suppression signal sent from the delay circuit can be subtracted from the inputted voice signal for activating the intercom function on which the ringing tone signal is superimposed, and sent to the voice recognition processing circuit. Therefore, the voice recognition processing circuit is a voice signal for activating the intercom function, which is superimposed on the ringing tone signal and input to the parent unit microphone when the ringing tone signal is output from the parent unit speaker. Can be recognized.

The intercom system according to the second aspect of the present invention is connected to an entrance cordless handset having a handset microphone and handset speaker via any one of a wired transmission path and a wireless transmission path. And a room base unit having a base unit microphone and a base unit speaker for inputting and outputting audio signals for making a call between the handset microphone and the handset speaker of the entrance handset. The living room master unit analog / digital converts the transmission analog voice signal input to the base unit microphone into a transmission digital voice signal, and is input to the slave unit microphone of the entrance unit and transmitted via the transmission path. A master audio codec that digitally / analog converts the incoming digital audio signal into an incoming analog audio signal and outputs it from the parent speaker, and a transmitted digital audio signal sent from the parent audio codec to the transmission path and the parent from the transmission path And an audio processing unit that performs transmission / reception control on the received digital audio signal transmitted to the machine audio codec. The voice processing unit generates a ringing tone signal to be output from the main unit speaker when the power is turned on and at the time of calling , and sends the ringing tone signal to the main unit voice codec. A sound recording circuit for storing a ringing tone signal transmitted after being converted from analog to digital via a parent device voice codec after being acoustically fed back from the parent device speaker to the parent device microphone after digital / analog conversion, and recorded at the time of calling A delay circuit that sends out a ringing tone signal read from the circuit after being delayed by an acoustic feedback time from the parent speaker to the parent microphone, and a ringing tone signal generation circuit that is generated at the time of calling through the parent unit voice codec. After the acoustic feedback from the main unit speaker to the main unit microphone, the interphone function with the ringing tone signal transmitted via the main unit audio codec is activated. From a subtracting circuit for subtracting the ringing tone signal sent from the delay circuit from the audio signal for the purpose, and a voice signal for activating the interphone function on which the ringing tone signal is superimposed based on the output signal from the subtracting circuit A speech recognition processing circuit for recognizing the speech signal.

  According to the interphone system of this aspect, when the system is turned on, the ringing tone signal to be output from the parent unit speaker of the living room base unit is generated by the ringing tone signal generation circuit of the voice processing unit. The sound can be stored in the recording circuit after being acoustically returned from the parent speaker to the parent microphone. Further, at the time of calling the system, the delay circuit delays the ringing tone signal read from the recording circuit by an acoustic feedback time from the parent speaker to the parent microphone and sends it to the subtracting circuit. Subtracting the ringing tone signal sent from the delay circuit from the voice signal for activating the intercom function on which the ringing tone signal is superimposed, which is input to the main unit microphone at the time of calling, and sends it to the voice recognition processing circuit Can do. Therefore, the voice recognition processing circuit is a voice signal for activating the intercom function, which is superimposed on the ringing tone signal and input to the parent unit microphone when the ringing tone signal is output from the parent unit speaker. Can be recognized.

  In the interphone system according to the third aspect of the present invention, in the first aspect or the second aspect of the present invention, the speech recognition processing circuit generates an acoustic model of a speech signal for activating the intercom function. The acoustic model generation function is such that when the acoustic model is generated, environmental dependency is formed using a sine wave having the same frequency as the ringing tone signal as background noise.

  In the intercom system of this aspect, the speech recognition processing circuit has an environment dependency in which the acoustic model of the speech signal for activating the intercom function has a background noise as a sine wave having the same frequency as the ringing tone signal. Can be generated. Therefore, the voice recognition processing circuit is a voice signal for activating the intercom function, which is superimposed on the ringing tone signal and input to the parent unit microphone when the ringing tone signal is output from the parent unit speaker. Can be recognized at a high level appropriately and reliably.

  According to the intercom system of the present invention, the intercom function is provided in which the ringing tone signal is superimposed and input to the base unit microphone at the time of calling the system in which the ringing tone signal is output from the base unit speaker of the living room base unit. It is possible to ensure the recognition accuracy of the audio signal for making the signal active appropriately and reliably at a high level corresponding to the environment dependency such as background noise.

The block diagram which shows the system configuration | structure of the intercom system by the 1st, 2nd Example of this invention. (A) The block diagram which shows the specific structure of the audio | voice processing part of a living room main | base station in the intercom system by the 1st Example of this invention. (B) The block diagram which shows the specific structure of the audio | voice processing part of a living room main | base station in the intercom system by the 2nd Example of this invention.

  Embodiments to which the intercom system of the present invention is applied will be described below with reference to the drawings.

  FIG. 1 is a block diagram showing a system configuration of an intercom system according to first and second embodiments of the present invention. In this intercom system, for example, an entrance cordless handset 1 installed outside a dwelling unit, such as a dwelling unit entrance, where a visitor calls a resident in the dwelling unit to make a call, and installed inside the dwelling unit, a wired transmission path L1 and wireless Residential base units 2a and 2b are provided in which a resident who has responded to a call from the entrance cordless handset 1 connected via any one of the transmission paths L2 communicates with a visitor. ing. Here, one of the room master units 2a and 2b is assumed to be the first room master unit 2a applied in the first embodiment. The other living room parent device 2b is the second living room parent device 2b applied in the second embodiment.

  Further, the entrance slave unit 1 shown in the figure includes a slave unit operation unit 10, a camera 11, a slave unit video processing unit 12, a slave unit microphone 13, a slave unit speaker 14, a slave unit audio codec 15, and a slave unit power supply unit 16. And the subunit | mobile_unit I / F (interface) 17 is provided.

In this entrance child device 1, the child device operation unit 10 is operated by a visitor to call a resident, and is configured by, for example, a call button, and can generate a ringing tone signal by the operation.

  The camera 11 generates an image signal by capturing an image of a visitor who has performed a call operation and an image (including a monitoring image) in the vicinity of the outside of the dwelling unit (the entrance to the dwelling unit). It is comprised with various video imaging media. Moreover, the subunit | mobile_unit image | video process part 12 performs predetermined signal processing, such as FM modulation and amplification, about the video signal produced | generated with the camera 11. FIG.

  The subunit | mobile_unit microphone 13 and the subunit | mobile_unit speaker 14 input / output an analog audio | voice (transmission audio | voice, reception audio | voice) signal, when a visitor talks with a resident. The handset audio codec 15 performs analog / digital conversion (A / D conversion) of an analog voice signal, which is a transmission voice input to the handset microphone 13, into a digital voice signal, and the first and second rooms. A digital audio signal which is a received voice that is input to the parent microphones 22a and 22b (to be described later) of the parent devices 2a and 2b and transmitted via any one of the wired transmission path L1 and the wireless transmission path L2. Is converted into an analog audio signal by digital / analog conversion (D / A conversion) and output from the handset speaker 14.

  Note that an amplifier (not shown) that amplifies an analog voice signal that is a transmission voice input to the handset microphone 13 can be provided between the handset microphone 13 and the handset codec 15. Further, an amplifier (not shown) that amplifies an analog voice signal that is a reception voice output from the handset speaker 14 can also be provided between the handset codec 15 and the handset speaker 14.

  The slave unit power supply unit 16 receives a power supply for a slave unit, which will be described later, supplied from the first and second living room master units 2a and 2b via the wired transmission line L1, and the configuration of the slave unit Power is supplied to each unit or received from the first and second living room master units 2a and 2b via a wireless transmission line L2, for example, power supplied by a non-contact power supply, which will be described later. It is supplied to each component of the slave unit.

  The slave unit I / F 17 includes a signal transmission line from the slave unit operation unit 10 to any one of the wired transmission path L1 and the wireless transmission path L2, and the wired transmission path L1 from the slave unit image processing unit 12. And a signal transmission line to any one of the wireless transmission lines L2 and a signal transmission line between the handset audio codec 15 and the wired transmission line L1, or the handset voice codec 15 and the wireless transmission line L2. And a power supply line from any one of the first transmission line L1 and the second transmission line L2 to the handset power supply unit 16 are respectively formed.

  Further, the first and second living room master units 2a and 2b shown in the figure are monitors 20a and 20b, master unit video processing units 21a and 21b, master unit microphones 22a and 22b, master unit speakers 23a and 23b, and a master unit. Audio codecs 24a and 24b, audio processing units 25a and 25b, master unit operation units 26a and 26b, master unit power supply units 27a and 27b, slave unit power supply units 28a and 28b, CPUs (master CPUs) 29a and 29b, and Master units I / F (interfaces) 30a and 30b are provided.

  In the living room master units 2a and 2b, the monitors 20a and 20b are controlled by the CPUs 29a and 29b (in a state where image output is possible), and a call display indicating that there is a call from the entrance slave unit 1 by a visitor, Each of them performs video display for displaying (outputting) a video signal generated by the camera 11 of the slave unit 1, and is composed of various video display media such as an LCD and an organic EL display. The master image processing units 21a and 21b are controlled by the CPUs 29a and 29b, and are transmitted from the entrance slave unit 1 via any one of the wired transmission path L1 and the wireless transmission path L2. For example, the video signal is subjected to predetermined signal processing such as FM demodulation and amplification, and the video signal generated by the camera 11 of the front door 1 is displayed (output) on the monitors 20a and 20b.

  The base unit microphones 22a and 22b and the base unit speakers 23a and 23b are used for inputting and outputting analog voice (transmitted voice and received voice) signals for a resident to make a call with a visitor. Of the machine speakers 23a and 23b, the master machine speaker 23a applied in the first embodiment is not only for the above-mentioned call function, but also when the system is called, that is, when the visitor calls from the front door 1 And a call notification function for outputting a ringing tone (ringing sound signal) indicating that there is a call, and when the system is powered on, that is, when power is supplied from the parent device power supply unit 27a to each component of the parent device. And a test notification function for outputting a test signal to be described later. On the other hand, the main unit speaker 23b applied in the second embodiment has not only the above-mentioned call function but also a call notification function for outputting a ringing tone (ringing tone signal) when the system is called and when the power is turned on. Is provided.

  The master unit audio codecs 24a and 24b are controlled by the CPUs 29a and 29b, and convert analog audio signals, which are transmission voices input to the master unit microphones 22a and 22b, into digital audio signals. Digital / analog conversion of a digital audio signal, which is a received voice input to the cordless handset microphone 13 and transmitted via any one of the wired transmission path L1 and the wireless transmission path L2, into an analog audio signal Thus, the data is output from the parent speaker 23a, 23b.

  In addition, between the master microphones 22a and 22b and the master codecs 24a and 24b, an analog voice signal that is a transmission voice input to the master microphones 22a and 22b, a ringing tone signal, or a test signal described later is superimposed. An amplifier (not shown) for amplifying an audio signal for activating the interphone function can be provided. Also, between the master codecs 24a and 24b and the master speakers 23a and 23b, an analog voice signal, a ringing tone signal, or a test signal described later, which is a received voice output from the master speakers 23a and 23b, is amplified. An amplifier (not shown) can be provided.

  The audio processing units 25a and 25b are controlled by the CPUs 29a and 29b, and are transmitted digital audio signals sent from the master audio codecs 24a and 24b to any one of the wired transmission line L1 and the wireless transmission line L2. In addition, transmission / reception control is performed on the received digital audio signal transmitted from either one of the wired transmission path L1 and the wireless transmission path L2 to the master audio codecs 24a and 24b.

  The base unit operation units 26a and 26b are detected by the CPUs 29a and 29b, and a resident who has confirmed that there is a call from the front cordless handset 1 by the visitor responds to the base unit microphones 22a and 22b and the base unit speaker 23a. , 23b to perform a response operation for making a call, or an end call operation to end the call, for example, a call button or a touch panel (not shown) provided in front of the monitors 20a and 20b ) Is preferred.

  The main unit power supply units 27a and 27b rectify commercial power and generate DC power. The slave unit power supply units 28a and 28b are controlled by the CPUs 29a and 29b and are used for operating the entrance slave unit 1 based on the DC power generated by the master unit power supply units 27a and 27b. It generates power.

  CPU29a, 29b controls each structure part of 1st, 2nd living room main | base station 2a, 2b. The master I / Fs 30a and 30b include a signal transmission line from any one of the wired transmission path L1 and the wireless transmission path L2 to the parent video processing units 21a and 21b, and the wired transmission path L1. And a signal transmission line between the base unit processing units 25a and 25b, or a signal transmission line between the wireless transmission path L1 and the voice processing units 25a and 25b, and the first transmission from the power supply units 28a and 28b for the slave units. A power supply line to any one of the line L1 and the second transmission line L2, a signal transmission line between the wired transmission line L1 and the CPUs 29a and 29b, or a wireless transmission line L2 and the CPUs 29a and 29b. And a signal transmission line between them.

  Next, a specific configuration of the voice processing unit 25a of the first room master unit 2a will be described with reference to the block diagram of FIG. The voice processing unit 25a includes a test signal generation circuit 250a, a recording circuit 251a, a transfer function calculation circuit 252a, a ringing tone signal generation circuit 253a, a suppression signal generation circuit 254a, a delay circuit 255a, a subtraction circuit 256a, a voice recognition processing circuit 257a, and A call control circuit 258a is provided.

In the audio processing unit 25a, the test signal generation circuit 250a generates a test signal (for measurement), for example, a time stretched pulse (TSP) signal to be output from the master speaker 23a when the system is turned on. Is sent to the audio codec 24a. In addition, the recording circuit 251a is subjected to digital / analog conversion via the base unit audio codec 24a, and then is acoustically returned from the base unit speaker 23a to the base unit microphone 22a and is analog / digital converted via the base unit audio codec 24a. After that, the transmitted test signal is stored, and is composed of various storage media such as RAM and EEPROM, for example. The transfer function calculation circuit 252a calculates an impulse response based on the test signal generated by the test signal generation circuit 250a and the test signal stored in the recording circuit 251a. The ringing tone signal generation circuit 253a generates a ringing tone signal to be output from the base unit speaker 23a when the system is called, and sends it to the base unit voice codec 24a. The suppression signal generation circuit 254a suppresses the acoustic feedback data of the ringing tone signal by convolving the impulse response calculated by the transfer function calculation circuit 252a with the ringing tone signal generated by the ringing tone signal generation circuit 253a. For generating a suppression signal. The delay circuit 255a delays the suppression signal generated by the suppression signal generation circuit 254a by an acoustic feedback time from the parent speaker 23a to the parent microphone 22a and sends the delayed signal to the subtraction circuit 256a. Further, the subtracting circuit 256a is generated by the ringing tone signal generating circuit 253a at the time of the above-mentioned calling and is acoustically fed back from the parent speaker 23a to the parent microphone 22a via the parent device voice codec 24a, and then the parent device voice codec 24a. The suppression signal sent from the delay circuit 255a from the voice signal for activating the intercom function superimposed on the ringing tone signal transmitted via the voice signal, for example, the voice signal used as the operation voice such as “Yes” Is subtracted. Furthermore, the voice recognition processing circuit 257a recognizes the voice signal from the voice signal for activating the interphone function on which the ringing tone signal is superimposed based on the output signal from the subtraction circuit 256a. An acoustic model generation function for generating an acoustic model of an audio signal for activating the function is provided. In this acoustic model generation function, when an acoustic model is generated, an environment dependency is formed in which a sine wave having the same frequency as the ringing tone signal is used as background noise.

  For example, the call control circuit 258a compares the sound pressure levels of input and output digital audio signals to form a signal transmission line for one audio signal having a high sound pressure level, and a signal transmission line for the other audio signal. Is a voice switch function that shuts off, for example, a technique disclosed in Japanese Patent Application Laid-Open No. 2011-55058, etc., and the noise level outside the dwelling unit where the entrance cordless handset 1 is installed and the first Without depending on the level difference of the noise level in the dwelling unit in which the room master unit 2a is installed, the communication path switching between the transmission state and the reception state in the room base unit, and the parent corresponding to the level difference An attenuator function or the like that amplifies or attenuates the signal level of the output signal from the machine speaker 23a can be provided.

  Next, a specific configuration of the voice processing unit 25b of the second room master unit 2b will be described with reference to the block diagram of FIG. The voice processing unit 25a includes a recording circuit 251b, a ringing tone signal generation circuit 253b, a delay circuit 255b, a subtraction circuit 256b, a voice recognition processing circuit 257b, and a call control circuit 258b.

In the voice processing unit 25b, the ringing tone generation circuit 253b generates a ringing tone signal to be output from the parent speaker 23b when the system is turned on and when the system is called, and sends the ringing tone signal to the parent unit voice codec 24b. is there. In addition, the recording circuit 251b is subjected to digital / analog conversion via the master audio codec 24b, then is acoustically returned from the master speaker 23b to the master microphone 22b, and is converted to analog / digital via the master audio codec 24b. The ringing tone signal transmitted after being transmitted is stored, and is constituted by various storage media such as RAM and EEPROM, for example. The delay circuit 255b delays the ringing tone signal read from the recording circuit 251b at the time of the above-mentioned calling by an acoustic feedback time from the main unit speaker 23b to the main unit microphone 22b and sends it to the subtraction circuit 256b. . Further, the subtracting circuit 256b is generated by the ringing tone signal generating circuit 253b at the time of the above-mentioned calling and is acoustically fed back from the parent speaker 23a to the parent microphone 22a via the parent device voice codec 24a, and then the parent device voice codec 24a. The ringing tone transmitted from the delay circuit 255b from the voice signal for activating the intercom function superimposed on the ringing tone signal transmitted via the voice signal, for example, the voice signal used as the operation voice such as “Yes” The signal is subtracted. Further, the voice recognition processing circuit 257b recognizes the voice signal from the voice signal for activating the intercom function on which the ringing tone signal is superimposed based on the output signal from the subtraction circuit 256b. Similar to the speech recognition processing circuit 257a applied in the first embodiment, an acoustic model generation function for generating an acoustic model of an audio signal for activating an intercom function on which a ringing tone signal is superimposed is provided. . In this acoustic model generation function, when an acoustic model is generated, an environment dependency is formed in which a sine wave having the same frequency as the ringing tone signal is used as background noise.

  Similar to the call control circuit 258a applied in the first embodiment, for example, the call control circuit 258b compares the sound pressure levels of input and output digital audio signals, and compares the sound pressure level of the one with the higher sound pressure level. A voice switch function that forms a signal transmission line for an audio signal and shuts off the signal transmission line for the other audio signal, or a technique disclosed in, for example, Japanese Patent Application Laid-Open No. 2011-55058. State of transmission in the room master unit without depending on the level difference between the noise level outside the unit where the machine 1 is installed and the noise level inside the unit where the second room master unit 2b is installed Or, it can be equipped with an attenuator function that amplifies or attenuates the signal level of the output signal from the main unit speaker 23b in response to the switching of the communication path in the receiving state and the level difference. Can.

  In the intercom system according to the first and second embodiments of the present invention configured as described above, specific operations will be described below.

  First, as a common operation in the first and second embodiments of the present invention, the master unit power supply units 27a and 27b of the first and second room master units 2a and 2b shown in FIG. It is possible to generate DC power obtained by rectifying AC power when power is turned on, and to supply power to each component of the parent device. The slave unit power supply circuits 28a and 28b generate a slave unit power source for operating the entrance slave unit 1 based on the DC power generated by the master unit power supply units 27a and 27b. Power is supplied from the I / Fs 30a and 30b to the slave unit power supply unit 16 via any one of the wired transmission line L1 and the wireless transmission line L2 and the slave unit I / F 17 of the entrance slave unit 1. The slave unit power supply unit 16 can supply the received slave unit power source to each component of the slave unit.

  Next, in the first embodiment of the present invention in which the first living room master unit 2a provided with the voice processing unit 25a shown in FIG. 2 (A) is provided, FIG. The CPU 29a of the first living room master unit 2a shown in the figure controls the master unit audio codec 24a to be active and outputs it from the master unit speaker 23a when controlling the voice processing unit 25a to be active (measurement). Test signal, for example, a time stretched pulse (TSP) signal, is generated by the test signal generation circuit 250a. This test signal is sent to the transfer function calculation circuit 252a, converted to digital / analog by the base unit audio codec 24a, output from the base unit speaker 23a, and then acoustically returned to the base unit microphone 22a. The test signal input to the parent microphone 22a as such acoustic feedback data is analog / digital converted by the parent voice codec 24a and then stored in the recording circuit 251a.

  Further, in the voice processing unit 25a of the first room master unit 2a, the transfer function calculation circuit 252a includes a test signal generated by the test signal generation circuit 250a and a test signal that is acoustic feedback data stored in the recording circuit 251a. Based on the above, an impulse response is calculated and sent to the suppression signal generation circuit 254a.

Thereafter, in order to call a resident in the dwelling unit in which the first living room main unit 2a is installed, a visitor outside the dwelling unit performs a predetermined calling operation using the child unit operation unit 10 of the front door unit 1. Then, the ringing tone signal generated by the slave unit operation unit 10 is transmitted to any one of the slave unit I / F 17, the wired transmission line L1, and the wireless transmission line L2, the first room master unit 2a. The data is transmitted to the CPU 29a via the parent device I / F 30a, and the camera 11 and the child device image processing unit 12 become active. In addition, the camera 11 starts capturing an image of a visitor who has performed a call operation (including an image in the vicinity of the vicinity outside the dwelling unit) and generates a video signal. This video signal is subjected to predetermined signal processing such as FM modulation and amplification in the slave unit video processing unit 12, and then any of the slave unit I / F 17, the wired transmission path L1, and the wireless transmission path L2 is selected. Is transmitted to the parent image processing unit 21a via the transmission path and the parent device I / F 29a.

When the CPU 29a of the first room master unit 2a receives the ringing tone signal transmitted from the entrance slave unit 1, it detects that there is a call from the entrance slave unit 1 by the visitor, and the monitor 20a and the master unit image are detected. The processing unit 21a is controlled to be active. The base unit video processing unit 21a performs predetermined signal processing such as FM demodulation and amplification on the video signal transmitted from the front unit 1 and sends it to the monitor 20a. The video signal generated by the camera 11, that is, the video of the visitor who performed the call operation (including video around the outside of the dwelling unit) is displayed (output) on the monitor 20 a.

  Further, the CPU 29a of the first room base unit 2a controls the base unit voice codec 24a to be active when the call is detected as described above, that is, when the system is called (see FIG. 2A). At the same time, the voice processing unit 25a is controlled to be active, and the ringing tone generation circuit 253a generates a ringing tone signal indicating that there is a call from the entrance cordless handset 1 by the visitor. This ringing tone signal is transmitted to the suppression signal generation circuit 254a, and after being digital / analog converted by the master unit voice codec 24a, is output from the master unit speaker 23a.

  Here, when the ringing tone signal is output from the master unit speaker 23a of the first room master unit 2a as described above, the resident who confirmed the video display on the monitor 20a together with such a call notification, When a voice signal for activating the interphone function, for example, a voice signal that is an operation voice such as “Yes”, is issued, the acoustic feedback data of the ringing tone signal activates the interphone function in the parent microphone 22a. Input in a state of being superimposed on the audio signal. A voice signal for activating the intercom function on which the ringing tone signal is superimposed, which is such a superimposed signal, is analog / digital converted by the base unit voice codec 24a, and then is sent to the subtracting circuit 256a of the voice processing unit 25a. Is transmitted.

  Further, in the voice processing unit 25a of the first room master unit 2a, the suppression signal generation circuit 254a applies the impulse response calculated by the transfer function calculation circuit 252a to the ringing tone signal generated by the ringing tone signal generation circuit 253a. By performing the convolution, a suppression signal for suppressing the acoustic feedback data of the ringing tone signal from the parent speaker 23a to the parent microphone 22a is generated and sent to the delay circuit 255a. The delay circuit 255a delays the suppression signal generated by the suppression signal generation circuit 254a by an acoustic feedback time from the parent speaker 23a to the parent microphone 22a and sends the delayed signal to the subtraction circuit 256a, whereby the subtraction circuit 256a The voice signal for activating the interphone function on which the ring tone signal from the base unit voice codec 24a is superimposed and the suppression signal transmitted from the suppression signal circuit 254a via the delay circuit 255a are synchronized at the input timing. Therefore, after the suppression signal is subtracted from the voice signal for activating the intercom function without causing a head break or the like, it is sent to the voice recognition processing circuit 257a.

  Furthermore, in the voice processing unit 25a of the first room master unit 2a, according to the voice recognition processing circuit 257a, a voice signal for activating the intercom function, which is an output signal from the subtraction circuit 256a, and a ringing tone signal The sound signal can be recognized appropriately and reliably with a high level of accuracy by comparing it with an acoustic model that has been generated in advance based on the environmental dependence of a sine wave having the same frequency as the background noise. Can do.

  Note that the voice signal for activating the intercom function recognized by the voice recognition processing circuit 257a in the voice processing unit 25a of the first room master unit 2a can be sent to the CPU 29a. According to the CPU 29a, for example, not only the main unit audio codec 24a and the audio processing unit 25a but also the sub unit audio codec 15 of the entrance sub unit 1 is activated, so that the sub unit microphone used by the visitor is used. 13 and handset speaker 14, handset voice codec 15, handset I / F 17, any one of first transmission path L 1 and second transmission path L 2, base unit I / F 30 a, base unit voice A signal transmission line is formed between the processing unit 25a (components of the processing unit), the base unit microphone 22a used by the resident and the base unit speaker 23a via the base unit audio codec 24a. Calls can be made by sending and receiving audio signals via the line. That is, the system is changed from a calling state to a calling state as an interphone function, and according to a resident, while confirming a video of a visitor displayed (displayed) on the monitor 20a continuously from the calling state. Can be called.

  As is apparent from the above description, according to the first embodiment of the present invention, at the time of calling the system in which a ringing tone signal is output from the parent device speaker 23a of the first room parent device 2a, The recognition accuracy of the voice signal for activating the intercom function, which is input to the main unit microphone 22a with this ringing tone signal superimposed, is appropriately and reliably set to a high level corresponding to the environmental dependence such as background noise. Can be secured.

  Next, in the second embodiment of the present invention in which the second living room master unit 2b including the voice processing unit 25b shown in FIG. 2B is provided, when the system is turned on as shown in FIG. The CPU 29b of the second room parent device 2b shown in the figure controls the ringing sound output from the parent device speaker 23b to control the parent device voice codec 24b to be active and the voice processing unit 25b to be active. The signal is generated by the ringing tone signal generation circuit 253b. Since this ringing tone signal is digital / analog converted by the base unit voice codec 24b, output from the base unit speaker 23b, and then acoustically returned to the base unit microphone 22b, the ringing tone signal which is such acoustic feedback data Is stored in the recording circuit 251b after analog / digital conversion by the master audio codec 24b.

Thereafter, in order to call a resident in the dwelling unit where the second living room main unit 2b is installed, a visitor outside the dwelling unit performs a predetermined calling operation using the child unit operation unit 10 of the front door unit 1. When this is done, as in the first embodiment described above, the ringing tone signal generated by the slave unit operation unit 10 is transmitted to any one of the slave unit I / F 17, the wired transmission line L1, and the wireless transmission line L2. The data is transmitted to the CPU 29b via the transmission path and the master unit I / F 29b of the second living room master unit 2b, and the camera 11 and the slave unit video processing unit 12 are activated. In addition, the camera 11 starts capturing an image of a visitor who has performed a call operation (including an image in the vicinity of the vicinity outside the dwelling unit) and generates a video signal. This video signal is subjected to predetermined signal processing such as FM modulation and amplification in the slave unit video processing unit 12, and then any of the slave unit I / F 17, the wired transmission path L1, and the wireless transmission path L2 is selected. This transmission path is transmitted to the master image processing unit 21b via the master I / F 29b of the second living room master 2b.

When the CPU 29b of the second room main unit 2b receives the ringing tone signal transmitted from the front door unit 1, the CPU 29b detects that the visitor has called from the front unit 1, and the monitor 20b and the main unit image are detected. The processing unit 21b is controlled to be active. The master unit video processing unit 21b performs predetermined signal processing such as FM demodulation and amplification on the video signal transmitted from the entrance slave unit 1 and then sends it to the monitor 20b, thereby sending the video signal of the entrance slave unit 1 to the monitor 20b. The video signal generated by the camera 11, that is, the video of the visitor who performed the call operation (including video around the outside of the dwelling unit) is displayed (displayed) on the monitor 20b.

  Further, the CPU 29b of the second living room master unit 2b performs control so that the master unit audio codec 24b is activated when the call is detected as described above, that is, when the system is called (shown in FIG. 2B). At the same time, when the voice processing unit 25b is controlled to be active, the ringing tone generation circuit 253b generates a ringing tone signal indicating that there is a call from the entrance cordless handset 1 by the visitor. The ring tone signal is digital / analog converted by the base unit voice codec 24b and then output from the base unit speaker 23b.

  Here, while the ringing tone signal is being output from the master speaker 23b of the second room master unit 2b as described above, the resident who has confirmed the video display on the monitor 20b together with such a call notification, As in the first embodiment, when a voice signal for activating the intercom function, for example, a voice signal that is an operation voice such as “Yes”, is sent to the parent microphone 22b, the acoustic feedback data of the ringing tone signal Are superimposed on the audio signal for activating the intercom function. A voice signal for activating the interphone function on which the ringing tone signal is superimposed, which is such a superimposed signal, is analog / digital converted by the base unit voice codec 24b, and then is sent to the subtracting circuit 256b of the voice processing unit 25b. Is transmitted.

  Further, in the voice processing unit 25b of the second room master unit 2b, the delay circuit 255b receives the ringing tone signal read from the recording circuit 251b at the time of system call (synchronized) from the master speaker 23b to the master microphone 22b. Is sent to the subtracting circuit 256b after being delayed by the acoustic feedback time, and the subtracting circuit 256b and the voice signal for activating the intercom function on which the ringing tone signal from the base unit voice codec 24b is superimposed (recording circuit) Since the ringing tone signal transmitted from the delay circuit 255b (from 251b) is synchronized at the input timing, the voice signal for activating the intercom function without causing a head break or the like is obtained. After subtracting the ringing tone signal, it is sent to the speech recognition processing circuit 257b.

  Further, in the voice processing unit 25b of the second room master unit 2b, according to the voice recognition processing circuit 257b, a voice signal for activating the interphone function, which is an output signal from the subtraction circuit 256b, and a ringing tone signal The sound signal can be recognized appropriately and reliably with a high level of accuracy by comparing it with an acoustic model that has been generated in advance based on the environmental dependence of a sine wave having the same frequency as the background noise. Can do.

  Note that the voice signal for activating the intercom function recognized by the voice recognition processing circuit 257b in the voice processing unit 25a of the second living room master unit 2b can be sent to the CPU 29b. According to the CPU 29b, as in the first embodiment described above, for example, not only the master unit audio codec 24b and the audio processing unit 25b but also the slave unit audio codec 15 of the front door unit 1 is made active. Thus, the slave microphone 13 and the slave speaker 14 used by the visitor, the slave voice codec 15, the slave I / F 17, the first transmission path L1, and the second transmission path L2 are selected. , Base unit I / F 30b, base unit voice processing unit 25b (components of each), signal transmission between base unit microphone 22b and base unit speaker 23b used by a resident via base unit audio codec 24b A line is formed, and a telephone call can be made by transmitting and receiving an audio signal via the formed signal transmission line. That is, the system is changed from a calling state to a calling state as an intercom function, and according to a resident, while confirming a visitor's image displayed (displayed) on the monitor 20b continuously from the calling state. Can be called.

  As is apparent from the above description, according to the second embodiment of the present invention, at the time of calling the system in which the ringing tone signal is output from the parent device speaker 23b of the second room parent device 2b, The recognition accuracy of the voice signal for activating the intercom function, which is input to the main unit microphone 22b with this ringing tone signal superimposed, is appropriately and reliably set to a high level corresponding to the environment dependency such as background noise. Can be secured.

  The intercom system of the present invention has been described with a specific embodiment. However, the present invention is not limited to this embodiment, and any intercom system of any configuration known so far, for example, a video, as long as the effects of the present invention are obtained. Interphone having the entrance cordless handset in which the imaging / signal processing functions (11, 12) are inadequate and the living room base unit in which the image output / signal processing functions (20a, 20b, 21a, 21b) are inadequate It goes without saying that even a system can be adopted.

DESCRIPTION OF SYMBOLS 1 ... Entrance cordless handset 13 …… Slave unit microphone 14 …… Slave unit speaker 2a …… First room base unit 22a …… Base unit microphone 23a …… Base unit speaker 24a …… Base unit voice codec 25a …… Sound Processing unit 250a …… Test signal generation circuit 251a …… Recording circuit 252a …… Transfer function calculation circuit 253a …… Ring tone signal generation circuit 254a …… Suppression signal generation circuit 255a …… Delay circuit 256a …… Subtraction circuit 257a …… Voice Recognition processing circuit 2b ...... second room master 22b …… master microphone 23b …… master speaker 24b …… master voice codec 25a …… speech processor 251b …… recording circuit 253b …… calling tone signal generation circuit 255b …… Delay circuit 256b …… Subtraction circuit 257b …… Speech recognition processing circuit L1 …… Wired transmission line L2 …… Wireless transmission line

Claims (3)

An entrance slave unit having a slave unit microphone and a slave unit speaker, and the slave unit microphone and the slave unit speaker of the entrance slave unit connected via any one of a wired transmission path and a wireless transmission path And a room base unit having a base unit microphone and a base unit speaker for inputting and outputting audio signals for making a call with
The living room base unit performs analog / digital conversion of a transmission analog voice signal input to the base unit microphone into a transmission digital voice signal, and is also input to the slave unit microphone of the entrance slave unit via the transmission path. The base unit voice codec that digitally / analog converts the received digital voice signal transmitted into the received analog voice signal and outputs it from the base unit speaker, and the transmission transmitted from the base unit voice codec to the transmission path A voice processing unit that performs transmission / reception control on the digital voice signal and the received digital voice signal transmitted from the transmission path to the base unit voice codec;
The voice processing unit generates a test signal to be output from the master speaker when the power is turned on, and sends the test signal to the master voice codec; and the digital / analog via the master voice codec a recording circuit for storing the test signal transmitted after the being analog / digital conversion is acoustically fed back via the base unit the audio codec to the base unit microphone from the master unit speaker after being converted, the test A transfer function calculating circuit for calculating an impulse response based on the test signal generated by the signal generating circuit and the test signal stored in the recording circuit; and a ringing tone signal to be output from the master speaker at the time of calling. A ringing tone signal generation circuit that generates and sends the ringing tone signal to the base unit voice codec, and the ringing tone generated by the ringing tone signal generation circuit A suppression signal generation circuit that convolves the impulse response calculated by the transfer function calculation circuit with the signal and generates a suppression signal for suppressing acoustic feedback data of the ringing tone signal, and the suppression signal generation circuit generates the suppression signal. A delay circuit for sending a suppression signal delayed by an acoustic feedback time from the parent speaker to the parent microphone, and generated by the ringing tone signal generation circuit at the time of calling, and via the parent unit voice codec, From the delay circuit, the voice signal for activating the intercom function superimposed on the ringing tone signal transmitted via the master unit voice codec after acoustic feedback from the master unit speaker to the master unit microphone is transmitted from the delay circuit. A subtracting circuit for subtracting the transmitted suppression signal, and the intercom function on which the ringing tone signal is superimposed based on an output signal from the subtracting circuit Intercom system, characterized in that it comprises a speech recognition processing circuit for recognizing the voice signal from the audio signal for the active. An entrance slave unit having a slave unit microphone and a slave unit speaker, and the slave unit microphone and the slave unit speaker of the entrance slave unit connected via any one of a wired transmission path and a wireless transmission path And a room base unit having a base unit microphone and a base unit speaker for inputting and outputting audio signals for making a call with
The living room base unit performs analog / digital conversion of a transmission analog voice signal input to the base unit microphone into a transmission digital voice signal, and is also input to the slave unit microphone of the entrance slave unit via the transmission path. The base unit voice codec that digitally / analog converts the received digital voice signal transmitted into the received analog voice signal and outputs it from the base unit speaker, and the transmission transmitted from the base unit voice codec to the transmission path A voice processing unit that performs transmission / reception control on the digital voice signal and the received digital voice signal transmitted from the transmission path to the base unit voice codec;
The voice processing unit generates a ringing tone signal to be output from the base unit speaker when the power is turned on and when calling and sends the ringing tone signal to the base unit voice codec, and the base unit voice when the power is turned on The ringing tone transmitted after the digital / analog conversion via the codec, the sound from the parent speaker to the parent microphone and the analog / digital conversion via the parent voice codec A recording circuit for storing a signal, a delay circuit for delaying and sending out the ringing sound signal read from the recording circuit at the time of the call by an acoustic feedback time from the parent speaker to the parent microphone, and at the time of the calling After being generated by the ringing tone signal generation circuit and being acoustically fed back from the parent device speaker to the parent device microphone via the parent device voice codec, A subtracting circuit for subtracting the ringing tone signal sent from the delay circuit from a voice signal for activating an intercom function superimposed on the ringing tone signal transmitted via a voice codec; And a voice recognition processing circuit for recognizing the voice signal from a voice signal for activating the interphone function on which the ringing tone signal is superimposed based on an output signal from the intercom system. The voice recognition processing circuit includes an acoustic model generation function for generating an acoustic model of a voice signal for activating the intercom function,
The environment dependence using the sine wave having the same frequency as the ringing tone signal as background noise is formed when the acoustic model generation function generates the acoustic model. 2. The intercom system according to 2.

Images