JP2003223179A - Intelligent loudspeaker - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an intelligent loudspeaker enabling easy control of voice output based on voice information given from an external controller. <P>SOLUTION: The intelligent loudspeaker is provided with a receiver section for receiving voice information including at least prescribed character data and phonological data, a voice synthesizing database, a voice data synthesizing section for generating synthetic voices corresponding to the voice information utilizing the voice synthesizing database, and a voice output section for outputting, based on the voice information, the generated synthetic voices. <P>COPYRIGHT: (C)2003,JPO

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an intelligent speaker, and more particularly to an intelligent speaker having a function of generating and outputting a synthetic voice based on received voice information.

[0002]

2. Description of the Related Art Conventionally, there has been provided a voice synthesizing device for preliminarily storing voice information such as phonemes and temperaments, converting input character data into voice and outputting the voice from a speaker.
Further, in Japanese Patent Laid-Open No. 2000-231396, phonological data and temperament data of a voice are extracted from an input voice so that a synthesized voice can be output with a natural intonation or a gap such as a dialogue of a drama. There is described a serif data creating apparatus for outputting a natural synthesized voice by using the extracted data, the unit voice data and the unit voice parameters of a plurality of speakers stored in advance.

Further, in Japanese Unexamined Patent Publication No. 8-249343, the input real-time voice is stored as digital voice data, and the voice of a part corresponding to a pre-registered wording pattern peculiar to news is stored in the digital voice data. There is described a voice information acquisition device that extracts data, further performs voice recognition, converts the data into text data, and efficiently searches information in an external database based on the text data.

Further, as shown in FIG. 3, there is also provided a private broadcasting system in which the same synthesized voice is output from the broadcasting amplifier 51 to a large number of speakers 52 to 63 by using a prerecorded synthesized voice. .

In the system of FIG. 3, the sound output from the broadcasting amplifier 51 is an analog signal, and the speaker converts the sent analog signal into sound and outputs the sound. Further, although FIG. 3 shows a system in which the speakers are divided into three groups, in this case, the broadcasting amplifier 51 is used.
Therefore, three different types of synthesized speech can be output for each group.

[0006]

However, Japanese Patent Laid-Open No. 2000-2000
Japanese Patent Laid-Open No. 231396 and Japanese Patent Application Laid-Open No. 8-249343 are for inputting a live human voice, and creating synthesized voice together with the information obtained from this voice, or performing information retrieval. , When outputting synthetic speech to a large number of speakers, it is possible to output natural speech or to output synthetic speech found by searching, but change the synthetic speech output for each speaker. Or
You cannot control the output time, number of times, and speaker selection.

Further, in the conventional system shown in FIG. 3, it is premised that the synthetic voice output to one group is output from all the speakers in the group,
It was not possible to control whether or not to output on the speaker side.

Therefore, the present invention has been made in consideration of the above circumstances, and a speaker is provided with functions such as voice synthesis and voice output control, and a signal given from an external control device is supplied. It is an object of the present invention to provide an intelligent speaker that can easily control the output of synthetic speech.

[0009]

According to the present invention, a receiving unit for receiving voice information including at least predetermined character data and phoneme data, a voice synthesis database, and the voice synthesis database are used to deal with the voice information. The present invention provides an intelligent speaker including a voice data synthesizing unit that generates a synthetic voice and a voice output unit that outputs the generated synthetic voice based on voice information.

According to this, since the synthesized voice is output by using the received voice information, the desired voice information is given from the external device so that the voice having the desired content can be transmitted at the desired time, the number of times, and the like. It can be output at a speed.
The voice data synthesizer is a given sentence, word, character,
This is a part that synthesizes so-called digital data such as symbols into information in a format that can be converted into analog voice according to a predetermined rule.

The speech synthesis database means words, characters,
It stores phoneme information for each symbol and basic information about phonemes, prosody, language, gender, etc. By combining these phoneme information and basic information, speech synthesis is performed. The voice information is created mainly by an external terminal device having a housing different from that of the intelligent speaker and arranged at a different position, and transmitted from the external terminal device to the intelligent speaker via a predetermined communication line.

The communication line is not particularly limited as long as it can transmit voice information as digital data.
It may be a wired line such as a LAN line such as 10BASE-T or a dedicated line directly connected, or a wireless communication line. Therefore, the receiving unit may have a receiving function applicable to the communication line and communication protocol used.

In a narrow sense, the voice output unit means a so-called speaker that converts digital data corresponding to voice data into an analog signal and outputs it as sound. In a broad sense, the voice output unit includes a receiving unit and a voice data synthesizing unit. It means a part that controls and interprets the received voice information and controls the overall output until the speaker outputs a sound. For overall output control, a microcomputer including a CPU, ROM, RAM, I / O controller, timer and the like can be used.

Further, the processing of each functional unit is performed by the ROM and the RAM.
It is executed by the operation of the CPU based on the program stored in the above. Speech synthesis database, C
Data required for PU operation are hard disk, R
It is stored in AM, MO, DVD-RAM or the like. Also,
Fixed data such as voice synthesis database is CD-RO
Those stored in M, CD-R, DVD-ROM, IC memory cards and the like can also be used.

The audio information of the present invention further includes output control data including any one of the number of output repetitions, the designated output time, the volume and the reading speed, and the audio output unit outputs the audio based on the control data. You may
For example, when the control data including the volume is received, the voice having the volume corresponding to the volume is output. By sending audio information including the specified output time to the intelligent speaker, the intelligent speaker itself can automatically control the output of the sound. Therefore, the external device that gives the control data always uses the intelligent speaker. There is no need to monitor and control, and the burden on the external device side can be reduced.

Further, the apparatus further comprises a number setting section for setting an identification number for identifying itself, and the voice information further includes a unique identification number for identifying the intelligent speaker, and the identification number set by the number setting section. If the identification number included in the voice information matches, the voice output unit may output the voice using the received voice information. According to this, even if a plurality of intelligent speakers are connected on a single communication line, it is possible to determine whether or not to output audio for each intelligent speaker and output audio. Therefore, different sounds can be output for each of the plurality of intelligent speakers connected to the single communication line.

Further, when the voice information includes a punctuation mark, the voice output unit may insert the predetermined pause period into the output timing of the punctuation mark without outputting the punctuation mark as a voice. . According to this, the sentence is separated for each punctuation mark and the voice is output, so that the output can be easily heard at a natural speed. In addition, even if a sentence to be output is a long sentence and is received after being divided into several pieces of audio information, if the punctuation marks are included in appropriate positions, the divided audio information will be output in combination. However, it is possible to output audio that is easy to hear in a natural tone that includes an appropriate pause period.

Further, a display unit is further provided, and when the received voice information includes a language designation or characters or symbols that cannot be output as voice, the character data included in the voice information is displayed on the display unit. Good. According to this
Even if it is not possible to output as a synthesized voice, such as when a language that is not stored in advance in the voice synthesis database is designated, or when the voice information includes symbols that cannot be pronounced, instead of voice output, By displaying the characters, the contents can be easily visually confirmed.

Further, according to the present invention, a plurality of intelligent speakers as described above and a control terminal for generating the voice information are connected via a communication line, and the control terminal has an intelligent speaker to output voice. The internal speaker broadcast system is characterized in that each intelligent speaker receives only the voice information in which the received voice information includes its own identification number, and outputs the voice information. It is provided. According to this, it is possible to individually control the availability of audio output for a plurality of intelligent speakers. Also, by including control data related to voice output in the voice information, the output control can be automatically performed on the intelligent speaker side, so that the control terminal side does not constantly monitor and control each intelligent speaker, thus reducing the burden. Can be reduced.

Here, the control terminal sets the designated output time for each intelligent speaker so as not to generate an echo in consideration of the relative distance between the plurality of intelligent speakers, and attenuates the loudness of the sound. The minimum volume to compensate for each speaker and minimize the interaction between the speakers is set for each intelligent speaker, and the specified output time for each intelligent speaker is set.
Audio information including volume is transmitted, and each intelligent speaker outputs at the specified output time included in the received audio information.
You may make it output the audio | voice data contained in the received audio | voice information. According to this, it is possible to prevent an echo from being heard due to a delay in the arrival time of the sound emitted from each intelligent speaker and an unnecessarily large output from the speaker.

Further, according to the present invention, a plurality of home control terminals and one or more intelligent speakers as described above are connected via a communication line, and each of the home control terminals has its control terminal. Voice information including message data peculiar to the user and the identification number of the intelligent speaker to be output as voice is transmitted, and the intelligent speaker receives and receives only the voice information including its own identification number. The present invention provides a home network system characterized by outputting message data included in voice information as voice. According to this, when a home control terminal is incorporated in various electric devices (for example, air conditioners, TVs, rice cookers) and an intelligent speaker is incorporated in a home controller or the like, message data transmitted from the various electric devices can be stored in the home controller. The voice can be output from the intelligent speaker of, and the state of electrical equipment can be intensively heard.

[0022]

BEST MODE FOR CARRYING OUT THE INVENTION The present invention will be described in detail below based on the embodiments shown in the drawings. The present invention is not limited to this. FIG. 1 is a block diagram showing the configuration of an embodiment of the intelligent speaker of the present invention.

In FIG. 1, reference numeral 11 is a voice data creating section included in a control device installed outside the intelligent speaker, and creates digital coded voice data having a structure as shown in FIG. 2 described later. It is the part to do. The control device installed externally corresponds to the broadcasting controller 70 represented by a personal computer or the like as shown in FIG. Reference numeral 1 in FIG. 1 denotes an external input terminal of the intelligent speaker, which is a portion for receiving the voice data sent from the voice data creation unit 11.

The intelligent speaker and the voice data creating section 11 are normally connected by one communication line. However, a device having a wireless communication function may be connected to the position of the external input terminal 1 so that data communication between the intelligent speaker and the voice data creation unit 11 is performed by wireless communication.

When a plurality of intelligent speakers are connected to one voice data creating section 11, each intelligent speaker may be connected by a different wire, but a LAN line of 10BASE-T specifications,
The connection may be made using a wireless LAN or the like. In this case, a device having a LAN connection function may be connected to the external input terminal 1 of the intelligent speaker.

The audio data receiving section 2 of FIG. 1 is a section for acquiring the audio data received via the external input terminal 1 in a predetermined data format shown in FIG. 2 as one unit. Audio data is stored in buffer memory 3
Stored in. As the buffer memory 3, a semiconductor memory such as RAM or a disk medium such as a hard disk can be used.

The ID number setting section 4 is for setting an identification number (ID number) unique to the intelligent speaker. In addition to the DIP switch and rotary switch, RO
The ID number recorded in M or the IC card may be read out. Alternatively, the setting data may be received via the external input terminal 1 and the ID number included in the setting data may be stored in the buffer memory 3. The timepiece drive unit 6 is a so-called built-in timepiece, and is used for outputting the current time, measuring a set time interval, and the like.

The voice data synthesizing unit 7 generates a synthesized voice by using the voice synthesizing database 8 based on the given voice information. The voice synthesis database 8 is
Digital information of a voice waveform corresponding to voice data such as characters and words is stored in advance. The speaker 9 converts the voice waveform data synthesized by the voice data synthesizer 7 into a sound wave and outputs the sound wave.

The display unit 10 is not an essential component from the viewpoint of broadcasting by voice, but
This is a visual display of setting information that should be displayed when using the intelligent speaker, data that cannot be converted into voice, or data that does not need to be voiced.
A display device such as LCD or EL display can be used. The display unit 10 can also be used for confirming the content of the sound broadcast by the speaker with characters.

The execution control section 5 is a section for operating each of the above functional sections, and can be composed of a microcomputer including a CPU, a ROM, a RAM, an I / O controller and the like. The execution control unit 5 executes the following functions, for example. (1) Storing the data received by the voice data receiving unit 2 in the buffer memory 3 (2) Reading the set ID number from the ID number setting unit 4 (3) Controlling the timepiece driving unit 6 (time setting, etc.) (4) Analysis of received voice data (5) Creation of data to be given to the voice data synthesizer 7 (6) Control of display 10 and creation of data to be displayed on the display 10.

In this intelligent speaker, the execution control unit 5 constantly monitors whether or not voice data is received via the voice data receiving unit 2, and the received voice data includes its own ID number. If it is one, it is stored in the buffer memory 3. Then, if the broadcast time is specified in the stored audio data, the execution control unit 5 confirms the current time obtained from the clock drive unit 6, and when the specified broadcast time is reached, the received audio data is received. Data obtained by processing the information contained in the data is given to the voice data synthesizing unit 7 to create a synthesized voice, and the voice is output from the speaker 9. More than,
It is a schematic operation of the execution control unit 5 of the intelligent speaker of the present invention.

FIG. 2 is a structural explanatory diagram of an embodiment of the audio data received by the audio data receiving section 2. One audio data starts with a header code 20 and ends with a termination code 29.
Ends with. Header code 20 and end code 29 are
These are special predetermined numerical data indicating the beginning and end of the audio data, respectively. The number 21 of destination IDs represents the total number of speakers performing broadcasting, and the destination ID number 22 represents the ID number of speakers performing broadcasting. Destination ID here
The number 22 is repeated by the number indicated by the destination ID number 21. That is, the destination ID number is not limited to one, but there may be a plurality of destination ID numbers.

The sender ID number 23 is an identification number that identifies the device that has transmitted the audio data, and is a number that identifies the broadcast controller 70, etc., which will be described later. The total number of bytes 24 is from the number of control data bytes 25 to the audio data 2
The total number of bytes of data up to 8 is shown.

The number of control bytes 25 is the total number of bytes of the control data 26. The control data 26 is information for controlling the output of the synthetic voice, and the voice data is output based on this information. For example, the control data 26 includes a “repeating code 30” indicating the number of repetitions, a “broadcast time code 31” designating the broadcast time, a “volume designating code 32” designating the volume, and a “reading speed designation” designating the reading speed. Code 33 "," Music designation code 34 "for designating the music data type such as MIDI of music and the music data body
Etc. are included.

The voice data byte number 27 is the voice data 2
8 total bytes. The voice data 28 is obtained by arranging a set of voice data A and a plurality of temperament data I (35 to 35).
42). The data received is shown in FIG.
In addition to voice data such as, data for controlling the intelligent speaker can be used. For example, it is possible to define the control data 26 that means time adjustment, and by transmitting this to all the intelligent speakers at the same time, the time adjustment of all the intelligent speakers is possible.

Next, FIG. 4 shows a block diagram of a private broadcasting system which is an embodiment of the present invention. This in-house broadcasting system can be used in schools, offices, factories, stations, stores, indoor and outdoor stadiums, meeting places, exhibitions, concerts and the like.

The local broadcast system shown in FIG. 4 comprises one broadcasting controller 70 and a plurality of intelligent speakers IS (71 to 82). The broadcasting controller 70 of FIG.
The audio data is transmitted. As the broadcasting controller 70, an information terminal such as a personal computer can be used.
The broadcasting controller 70 creates audio data having the configuration shown in FIG.

The voice data 28 in FIG. 2 can be input as character data from a keyboard provided in the broadcasting controller. Alternatively, a broadcast original printed on paper is read by a scanner and converted into character data by character recognition, or a voice input from a microphone is converted into character data by voice recognition. Further, in the broadcasting controller 70, prosody data is created for the input audio data, the control data 26 as shown in FIG. The voice data shown in 2 is created.

Although the broadcasting controller 70 and each intelligent speaker IS (71 to 82) can be connected to each other by different wires, for example, 10 BA for each floor and each predetermined installation area of the intelligent speaker.
You may connect by LAN lines, such as SE-T. Alternatively, the broadcast controller 70 may communicate with each intelligent speaker by wireless communication using infrared rays, light, ultrasonic waves, radio waves, or the like.

In FIG. 4, the ID number of the broadcasting controller 70 is "0-0" and the ID numbers of the intelligent speakers are "IS1-1" to "IS3-4".

FIG. 5 is an explanatory diagram of an embodiment of audio data used in the embodiment of this private broadcasting system. Here, the number of destination IDs 84 is 3, and subsequently, three destination ID numbers (84, 85, 86) are described. That is, by these numbers, this voice data is transmitted to the three intelligent speakers IS1-1, IS2-2, I.
Indicates that the data is sent to S1-3.

In FIG. 5, the voice data 28 is
Three character data of "a", "i", and "u" (92, 9
4, 96) and the prosody data (93, 95, 9)
7) is "0". Here, the prosody data being 0 indicates that there is no change from the standard setting prosody data, that is, the standard prosody prepared in advance is used. 3 which received the voice data shown in FIG.
Intelligent speakers IS1-1, IS2-
2, IS1-3 will output the voice corresponding to the character string "Ai" in standard prosody.

Next, FIG. 6 shows a flowchart of a process until the execution control unit 5 of the intelligent speaker outputs a voice using the received voice data. here,
It is assumed that the intelligent speaker is preset with either a “real time mode” or a “punctuation mark mode” as a voice output mode. This setting is, for example, the reading mode data 4 included in the control data.
3 can be used. Here, the “real-time mode” means a mode in which all the received voice data are output as voice at once. The "punctuation mark mode" means a mode in which "punctuation marks" or "reading marks" included in the received voice data are detected, and voice output is performed in a character string unit up to the punctuation marks or the reading marks with a pause period.

First, in step 100, voice data is received and stored in the reception buffer of the buffer memory 3. It is checked whether the destination ID number in the received data matches the set own ID number (step 101). If they do not match, the buffer memory 3
The reception buffer of is cleared (step 102) and the process returns to step 100.

If they match, it means that the broadcast data has been transmitted to itself, so that the process proceeds to step 103. In step 103, it is confirmed whether the received data includes the broadcast time data 31.

When the broadcast time is designated, the process proceeds to step 112, the time data is temporarily stored, and broadcast at the designated time. If the broadcast time is not designated, the process proceeds to step S104, and it is confirmed whether or not the repeated data 30 is present. If the repeat count is specified, the process proceeds to step 113, where the repeat count data is temporarily stored,
Broadcast a specified number of times. If the number of repetitions is not specified, the process proceeds to step 105 and it is confirmed whether the "punctuation mode or real-time mode" is currently set.

If the real-time mode is set, the process proceeds to step 110, where all the audio data stored in the buffer memory 3 is moved to the output buffer and given to the audio data synthesizer 7. Then, in step 111,
The voice data synthesizer 7 creates a synthesized voice and outputs the voice from the speaker 9.

In step 105, in the case of the "punctuation mode", the process proceeds to step 106, and only one voice data is read from the reception buffer of the buffer memory 3.
Here, one means one audio data (the audio data 28 of FIG.
1 character data in).

In step 114, if there is no data to be read, the process returns to step 100, and if there is voice data, the process proceeds to step 109. In step 109, the read audio data is added to the output buffer in the buffer memory 3. If audio data already exists in the output buffer, the audio data is saved after the audio data.

Next, in step 107, it is checked whether the added voice data is a punctuation mark, and in step 108 it is checked whether it is a reading mark. If it is not a punctuation mark or a reading mark in steps 107 and 108, the process returns to step 106 and the process is repeated for the next voice data.

If it is a punctuation mark or a reading mark, step 11
The process proceeds to 1 and the voice data stored in the output buffer is given to the voice data synthesizing unit 7 to output voice by voice synthesis. Here, at the time of voice synthesis, the punctuation marks are not output as voice, but are used to provide a pause period for a predetermined time. By providing the pause period in this way, it is possible to make the broadcast easy to hear.

FIG. 7 shows a flow chart of a process of judging whether or not voice data synthesis of the intelligent speaker of the present invention is possible. First, in step 120, whether or not there is “language designation data” in the control data 26 of the received voice data and the language designated language are included in the languages stored in the voice synthesis database 8. Confirm whether or not If the designated language is in the voice synthesis database 8, the process proceeds to step 121, where voice synthesis output is performed. Step 1 if no language is specified
In step 22, the output buffer data is displayed and output as character data on the display unit 10 without voice output.

FIG. 8 shows an embodiment of a private broadcasting system in which the same audio data is output from each intelligent speaker after being shifted by a predetermined time in consideration of the installation position of the intelligent speaker. In FIG. 8, IS1-1 (131), IS is placed at a position 60 m away from the intelligent speaker IS2-1 (134).
2-2 (135) and IS3-1 (137) are installed,
IS1-2 (132), IS2 at a position 120 m away
-3 (136) and IS3-2 (138) are installed, 1
IS1-3 (133) and IS3- at a distance of 80 m
3 (139) are installed.

Now, assuming that the sound velocity is 300 m / sec, IS2
The sound output from -1 reaches the position of 60 m after 0.2 seconds, reaches the position of 120 m after 0.4 seconds, and reaches 0.
The position of 180 m is reached after 6 seconds. If the same audio data is simultaneously output from all intelligent speakers, for example, at the position of IS2-2, IS2
-2, 0.2 seconds after voice data is output, IS2
The same audio data output from -1 will be heard, and it will be heard as if so-called echo is applied, and the broadcast will be difficult to hear.

Therefore, in order to mitigate the occurrence of this echo, the output timing of each intelligent speaker is adjusted in consideration of the arrival delay time and the attenuation amount of the voice and the sound volume.

FIG. 9 shows a case where the arrival delay time of voice is taken into consideration.
An example of one broadcast data is shown. Here, the broadcasted sounds are all the same. Broadcast data 1 is data given to IS2-1, broadcast data 2 is data given to IS1-1, 2-2, 3-1 and broadcast data 3 is IS1-2, 2-3, IS1-2. 3-2, and the broadcast data 4 is IS1-3.
This is the data given to 3-3.

The four broadcast data are time data 31.
And the volume is different from the time designated as the volume data 32. The time data of the broadcast data 1 is 18:15:30 on August 8, 2001, and the volume data is 100. The time data of the broadcast data 2 is set to 0.2 seconds later than this, and the volume data is set to be smaller by 24. That is, 0.2 seconds after the broadcast data 1 "Ai" broadcast from IS2-1, the volume is 76, IS1-1, 2-2, 3 installed 60 m away from IS2-1.
Broadcast data 2 "Ai" is designated to be broadcast from -1.

Similarly, the broadcast data 3 sent to IS1-2, 2-3, 3-2 installed 120 m away from IS2-1 includes time data and volume data 0.4 seconds later than broadcast data 1. 72 is set and broadcast data 4 sent to IS1-3 and 3-3 installed 180m apart
Is set with time data and volume data 71 which is 0.6 seconds later than the broadcast data 1.

If such four types of broadcast data are transmitted from the broadcasting controller 130 to each intelligent speaker, the sound output from the intelligent speaker and the intelligent speaker IS2-1 are output at the position of each intelligent speaker. Sound is heard at almost the same time, and the volume is heard at a volume of 100 in which the volume of each speaker is superimposed. Unnatural echo can be prevented and the volume can be prevented from being impaired.

If a "volume designating code" having predetermined volume data is also given to each intelligent speaker as control data at the same time, the volume can be adjusted.
You can make it easier to hear. As an example, FIGS. 12 and 13 show changes in volume in the sound propagation direction and changes in volume in the opposite direction. FIG. 12 shows changes in the volume of IS2-1, 2-2, 2-3 in the propagation direction of broadcasting. The sound of each intelligent speaker is adjusted such that the sound phase is adjusted at the broadcast time and the sound volume is always 100 at the speaker position in consideration of sound attenuation. Therefore, the sound of each intelligent speaker is synthesized with the same phase as a thick line, which is effective in making the broadcast easier to hear. FIG. 13 shows changes in the volume of sound transmitted in the direction opposite to the propagation direction of broadcasting for IS2-1, 2-2, 2-3. The sound of each intelligent speaker has a different sound phase when propagating in the reverse direction. For example, when the sound of another intelligent speaker is heard by IS2-1, the sound output by IS2-2 is heard 0.4 seconds later, and the sound of IS2-3 is heard 0.8 seconds later. In this way, the sounds of the intelligent speakers heard at the IS2-1 position are out of phase, and the volume of the intelligent speakers is set lower than that of general public broadcasting according to the volume data. Since it produces a single sound like a plurality of thick lines, it has an effect of suppressing the influence of echo smaller than that of general public broadcasting.

FIG. 10 shows an explanatory diagram of an embodiment in which the intelligent speaker of the present invention is used in a home network system. In FIG. 10, the home network system includes an electric kettle 180 as an electric device, an air conditioner 181, a refrigerator 182, a television 183, and a home controller 185. These devices are connected by a predetermined data transmission path 193. There is.

The data transmission path 193 may be a wired line such as a LAN or a power line, and a wireless communication section may be provided in each device for wireless communication. The intelligent speaker is provided inside the home controller 185. Unlike the broadcasting system shown in FIG. 1, this embodiment has a plurality of controllers and one intelligent speaker.
The audio data transmitted from each controller (180 to 183) is output from one speaker. In addition to the voice data of FIG. 2 described above, the data transmission path 193 is used to transmit control signals (start, stop, setting change, etc.) for controlling the electric devices (180 to 183) from the home controller 185. You may use it.

For example, when voice data is transmitted from the electric pot 180 of FIG. 10 to the home controller 185,
The sender ID number is 0-1 and the destination ID number is 1-1, and voice data including a predetermined message (such as "I will cook in 10 minutes") is sent and the message is output from the speaker of the home controller 185. To be done.

In addition, as a message sent from the air conditioner, "The temperature at room temperature has become the same as the outside temperature.
We recommend turning off the air conditioner. The message sent from the refrigerator says, "The refrigerator door is 1
It has been open for 0 minutes. And so on.

In the embodiment of FIG. 10, a message is sent from the electric device to the home controller 185 each time, so it is not necessary for the home controller side to store all messages corresponding to a plurality of connected electric devices. Since it does not exist, the configuration of the home controller can be facilitated. Further, even if a new electric device is connected, it is not necessary to newly register a message corresponding to the electric device in the home controller.

Unlike FIG. 10, FIG. 11 is a configuration diagram showing a home network system in which a plurality of home controllers (188, 189) and the intelligent speaker itself 190 are further connected. The home controller may be installed in each room, for example.

In the embodiment shown in FIG. 11, electric equipment (186,
187) sends the message to all home controllers 188, 189 and intelligent speaker 190 so that the user can hear the message anywhere in the home. In addition, if the destination of the message is set in advance to the electric device from the home controller side using the control signal, the message from the electric device can be selectively transmitted only to the set home controller.

[0068]

According to the present invention, since the synthesized voice is output by using the received voice information, the desired voice information is given from the external device so that the voice having the desired content can be output at the desired timing. It can be output at high speed. Especially, by transmitting control data such as specified output time to the intelligent speaker, the intelligent speaker itself can automatically control the output of the sound. It is not necessary to monitor and control the device, and the load on the external device side can be reduced.

[Brief description of drawings]

FIG. 1 is a configuration block diagram of an embodiment of an intelligent speaker of the present invention.

FIG. 2 is a block diagram of an embodiment of audio data of the present invention.

FIG. 3 is a configuration block diagram of a conventional private broadcasting system.

FIG. 4 is a configuration block diagram of an embodiment of a local area broadcasting system using the intelligent speaker of the present invention.

FIG. 5 is an explanatory diagram of a specific configuration of audio data used in the private broadcast system of the present invention.

FIG. 6 is a flowchart of the entire processing of the intelligent speaker of the present invention.

FIG. 7 is a flowchart of a process of determining whether to output or display voice according to the present invention.

FIG. 8 is an explanatory diagram of another embodiment of the private broadcast system of the present invention.

FIG. 9 is an explanatory diagram of broadcast data of another embodiment of the private broadcast system of the present invention.

FIG. 10 is a configuration block diagram of an embodiment of a home network system using the intelligent speaker of the present invention.

FIG. 11 is a configuration block diagram of an embodiment of a home network system using the intelligent speaker of the present invention.

FIG. 12 is a graph showing changes in volume in the sound propagation direction.

FIG. 13 is a graph showing changes in volume transmitted in a direction opposite to the sound propagation direction.

[Explanation of symbols]

1 External input terminal 2 Audio data receiver 3 buffer memory 4 ID number setting section 5 Execution controller 6 Clock drive 7 Voice data synthesizer 8 Speech synthesis database 9 speakers 10 Display 70 Broadcast controller

Front page continuation (51) Int.Cl. ⁷ Identification code FI theme code (reference) H04R 3/12 G10L 3/00 QS

Claims (8)

[Claims] 1. A receiving unit that receives voice information including at least predetermined character data and phoneme data, a voice synthesis database, and voice data synthesis that generates a synthesized voice corresponding to the voice information using the voice synthesis database. An intelligent speaker, comprising: a unit and a voice output unit that outputs a synthesized voice generated based on voice information. 2. A number setting unit for setting an identification number for identifying itself is provided, and the voice information further includes a unique identification number for identifying an intelligent speaker for outputting voice, and the number is set by the number setting unit. The intelligent speaker according to claim 1, wherein when the identification number and the identification number included in the voice information match, the voice output unit outputs voice using the received voice information. 3. The audio information is the number of output repetitions,
The intelligent speaker according to claim 2, further comprising output control data including any one of a designated output time, a volume, and a reading speed, and the voice output unit outputs a voice based on the control data. 4. The voice output unit, when the voice information includes a punctuation mark, does not output the punctuation mark as a voice and inserts a predetermined pause period at the output timing of the punctuation mark. The intelligent speaker according to claim 3. 5. A display unit is further provided, and when the received voice information includes a language designation or characters or symbols that cannot be output as voice, the character data included in the voice information is displayed on the display unit. The intelligent speaker according to claim 1. 6. The plurality of intelligent speakers according to claim 2 are connected to a control terminal for generating the voice information via a communication line, and the control terminal has an intelligent speaker to output voice. An internal broadcasting system characterized in that an identification number is transmitted by including it in voice information, and each intelligent speaker receives only voice information in which the received voice information includes its own identification number and outputs the voice. 7. The control terminal sets an output designated time for each intelligent speaker so that an echo does not occur in consideration of a relative distance between the plurality of intelligent speakers, and an output set for each intelligent speaker. 7. The internal broadcasting according to claim 6, wherein the intelligent speaker transmits audio information including a specified time, and each intelligent speaker outputs the audio data included in the received audio information at an output specified time included in the received audio information. system. 8. A plurality of home control terminals and one or more intelligent speakers according to claim 2 are connected via a communication line, and each of the home control terminals is connected to the control terminal. The voice information including the peculiar message data and the identification number of the intelligent speaker to be output as voice is transmitted, and the intelligent speaker receives only the voice information including its own identification number, and the received voice is received. A home network system characterized in that message data included in information is output as voice.