JPH0575482A - Voice transmitter - Google Patents

Abstract

PURPOSE:To send a voice signal automatically while being divided at a position easily listened to naturally by synthesizing divided characteristic data with voice rules, processing the data into voice data and sending the data in the unit of divisions in a radio wave. CONSTITUTION:Digital voice data with a length not in excess of a prescribed length recorded in 1st and 2nd recording sections 11a, 11b comprising, e.g. a semiconductor memory at a transmitter side 10 are read to a voice reproduction section 12 according to the sequence. The voice reproduction section 12 reproduces the read voice data by e.g. the ADM system and the obtained voice data are outputted to a radio transmission section 13. Furthermore, the radio transmission section 13 modulates the voice data and superimposes the modulated data onto a carrier with a specified frequency and sends the result as a radio wave. In this case, the voice data are divided in the grammatical unit such as blocks of sentences or paragraphs or the like not in excess of a prescribed length and they are recorded on 1st and 2nd recording sections 11a, 11b and the recorded voice data are reproduced and sent in the divided units.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a voice transmitting apparatus for transmitting voice data by using a radio apparatus which cannot perform continuous transmission for a certain time or longer.

[0002]

2. Description of the Related Art Speech recognition technology is an excellent man-machine
It plays an important role in realizing the interface. Then, as a method for the speaker to input voice to the voice recognition device while moving freely, the voice recognition device is miniaturized and built into the transmitter side, and the code signal of the synthesis result is skipped by wireless communication and input to the computer. And a method in which a voice signal input from a microphone by a speaker is skipped by wireless communication and input to a voice recognition device. At the same time, it is necessary to use a voice synthesizing device to synthesize an explanation for using the voice recognizing device and a message for the recognition result, and inform the speaker by voice over the wireless.

By the way, for example, "specific low power radio station"
When the (Radio Law Enforcement Regulations, Article 6) is applied to the above wireless communication, it is not possible to keep the carrier wave in the transmitting state at all times due to the limitation of the continuous communication time and the provision of the transmission suspension. Therefore, when it is desired to synthesize and transmit a message or the like that is longer in time than the limit of the continuous communication time, the transmission is stopped regardless of the content phrase of the message. The voice that is difficult to hear will be received.

[0004]

As described above, in the conventional case, when a wireless station having a limitation of continuous communication time or a transmission suspension regulation is used, it is longer in time than the limitation of continuous communication time. When a message or the like is speech-synthesized and transmitted, the message is suspended in the middle of the communication as the continuous communication time elapses without any context, so that a voice that is very difficult to hear is received.

The present invention has been made in view of the above circumstances, and an object of the present invention is to transmit a voice in a state in which it is easy to hear when it is expected that the voice transmission will be stopped midway. An object of the present invention is to provide a speech synthesizer capable of performing the above.

[0006]

SUMMARY OF THE INVENTION That is, the present invention is a voice transmitting device for reproducing and transmitting voice recorded by using a wireless device which cannot continuously transmit for a fixed time or longer, and does not exceed a fixed length. As described above, the voice data is divided into a plurality of predetermined units in the grammar, for example, breaks such as sentences and clauses, and the voice data is divided and recorded in advance. The recorded voice data is reproduced and transmitted for each division unit. I made it,
When the transmission of the reproduced voice is expected to be suspended midway, the voice can be divided and transmitted at natural and easily audible positions.

Further, the present invention is a voice transmitting device for transmitting voice data obtained by performing voice rule synthesis from a series of character data, using a wireless device which is not capable of continuous transmission for a certain period of time or more. The output time after rule composition is calculated, and in accordance with the calculated output time, the character data is divided into a plurality of predetermined units in the grammar, such as sentences and clauses, so that the character data does not exceed a certain time length that can be continuously transmitted. It is divided into two parts, and the divided character data is synthesized by voice rules and converted into voice data, and then wirelessly transmitted every time the above is divided.It is expected that the transmission of the rule-combined voice will be suspended midway. In this case, it is possible to automatically divide the voice at a position where it is natural and easy to hear and divide the voice for transmission.

[0008]

First Embodiment A first embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 shows a schematic circuit configuration thereof.
The range indicated by is the transmitter side, the range indicated by 20 is the receiver side, and here it is assumed that the transmitter side 10 performs wireless communication based on the "specified low power radio station" to the receiver side 20. ..

On the transmitter side 10, the audio data of digital values recorded in the first recording section 11a and the second recording section 11b, which are composed of, for example, a semiconductor memory, each having a length not exceeding a certain length, are arranged in the order. It is read by the audio reproduction unit 12. The audio reproduction unit 12 reproduces the read audio data by, for example, the ADM method, and outputs the obtained audio data to the wireless transmission unit 13. The wireless transmission unit 13 modulates voice data, superimposes it on a carrier wave of a prescribed frequency, and transmits it as a radio wave. In this case, the length of the audio data recorded in each of the recording units 11a and 11b is determined in consideration of the continuous transmission possible time in the wireless transmission unit 13. On the receiver side 20, the transmitted radio waves are received by the wireless reception unit 21, demodulated to obtain original voice data, and the voice output unit 22 outputs this.

In addition, as shown in FIG.
When it is desired to record a large amount of audio data such that the storage capacity is insufficient with only 11a and 11b, a recording unit may be further added like the nth recording unit 11n as shown in FIG. Next, the operation of the first embodiment will be described.

Now, let us say that, for example, the message "The water temperature is abnormal. Please check the cooling device" shown in FIG. 3 is output as voice. When the time limit for continuous transmission by the wireless transmission unit 13 is "T" seconds, the time required to output this message by voice is "T + A" seconds (T
> A), this message is 1
Since it cannot be transmitted within one transmission time, it is necessary to transmit in two times.

Here, simply as shown in FIG. 4, one message is recorded only in the first recording section 11a, transmitted for "T" seconds first, and remains after "S" seconds transmission is paused. When you send "A" seconds, the contents of the transmission are "Water temperature is abnormal. Check the cooling device" and "Please." As shown in the figure, and the transmission is paused between "shi" and "te". Since it is performed, the message is very difficult for the receiver side 20 to hear.

Therefore, as shown in FIG. 5, the above message is displayed for "B" seconds (B <T) "Water temperature is abnormal" and "C".
It is divided into two messages of "Please check the cooling device" for seconds (C <T) and recorded in the first recording section 11a and the second recording section 11b respectively, and first the first recording Department
After sending the message "Water temperature is abnormal." For "B" seconds recorded in 11a and suspending the transmission for "S" seconds, the "C" seconds recorded in the second recording unit 11b. Send the message "Check the cooling system." In this way, the voice data is divided into sentences on the grammar unit, and by inserting a transmission pause between the divided sentences, even if the message is divided into two and the message is transmitted,
The voice output unit 22 can output a natural and easily audible message as voice.

If the "C" seconds are longer than the "T" seconds and the transmission pause is inserted between the sentences, the second message is "point the cooling device" as shown in FIG.
In the case where it is further divided into two such as "Please check.", If it is transmitted as it is and output as voice, the message becomes very difficult to hear as in the case shown in FIG.

Therefore, as shown in FIG. 7, the above message is displayed as "E" seconds (E <T) "cooler" and "F" seconds (F <T) "check." It is divided into one message and recorded in the second recording section 11b and the third recording section 11c (not shown) respectively, and the "cooling device for" E "seconds recorded in the second recording section 11a is recorded. Message is sent and the transmission is paused for "S" seconds, then the third recording unit
Send the message "Please check." For "F" seconds recorded in 11c. In this way, the voice data is divided into phrases on the grammar unit, and by inserting a transmission pause between the divided phrases, even if the message is divided and transmitted, the voice output unit 22 naturally listens to it. A simple message can be output by voice.

In the first embodiment, the first recording section 11
Although it has been described that the a and the second recording unit 11b are configured with the semiconductor memory, the present invention is not limited to this, and other recording media such as a tape recorder may be used.

Similarly, although it has been stated that the voice reproducing section 12 reproduces from digital value voice data by the ADM method, the present invention is not limited to this, and other voice synthesizing methods such as ADPCM may be adopted. Of course.

[0019]

Second Embodiment Next, a second embodiment of the present invention will be described with reference to the drawings.

FIG. 8 shows a schematic circuit configuration thereof.
The range indicated by is the transmitter side, and the range indicated by 40 is the receiver side. Here, it is assumed that wireless communication is performed from the transmitter side 30 to the receiver side 40 based on the "specified low power radio station". ..

On the transmitter side 30, the text data consisting of a series of character code data input by the text data input unit 31 composed of, for example, a word processor is first read to the output time calculation unit 32.

The output time calculation unit 32 calculates the output time required for voice synthesis from the read text data according to predetermined voice synthesis information such as reading speed, and when the predetermined continuous transmission time is not exceeded, Outputs the sentence data to the voice synthesis unit 33 as it is, and outputs the calculated output time data and the sentence data to the sentence division unit 34 when exceeding.

The sentence division unit 34 divides the sentence data into grammatical sentence units according to the output time data from the output time calculation unit 32 and the sentence data, and the divided result is the voice synthesis unit 33 and the wireless transmission unit 35. To send to.

The voice synthesis unit 33 performs voice rule synthesis from the sentence data sent from the output time calculation unit 32 or the sentence division unit 34, for example, by the waveform synthesis method, and outputs the obtained voice data to the wireless transmission unit 35. To do.

The radio transmitter 35 modulates the voice data from the voice synthesizer 33 into a carrier wave of a specified frequency while inserting a transmission pause according to the transmission timing according to the division result input from the sentence divider 34. Superimpose and transmit as radio waves. On the receiver side 40, the radio wave received by the radio receiver 41
The audio output section 42 outputs the original audio data obtained by demodulation.

Next, the operation of the second embodiment will be described. The text data used for speech synthesis and the processing contents thereof are the same as those used in the first embodiment, and the above-described FIGS. 3 to 7 are used.

That is, it is assumed that the text data of the message "The water temperature is abnormal. Please check the cooling device" shown in FIG. If the limit is "T" seconds, the time required to voice output this message is "T +
If it is "A" seconds (T> A), this message cannot be transmitted within one transmission time as a matter of course, so it is necessary to transmit the message in two times.

Here, as shown in FIG. 4, the data read from the character data input unit 31 is first calculated by the output time calculation unit 32 for "T" seconds, and transmitted by the voice synthesis unit 33 and the wireless transmission unit 35. And "A" remains after pausing transmission for "S" seconds.
When sending seconds, as shown in the figure, the contents of the transmission are "Water temperature is abnormal. Check the cooling device" and "please." The transmission is paused between "shi" and "te". This makes the message very difficult to hear.

Therefore, in accordance with the output time calculation result of the output time calculation unit 32, the sentence division unit 34 detects the phrase ".", For example, and outputs the message "B" seconds (B) as shown in FIG.
<T) “The water temperature is abnormal.” And “C” seconds (C <T)
"Please inspect the cooling device." Is divided into two messages, and the two divided sentence data are output to the voice synthesis unit 33, and the output time data of "B" seconds and "C" seconds are wirelessly transmitted to the wireless transmission unit. Send it to 35, send the message "Water temperature is abnormal" for "B" seconds first, pause the transmission for "S" seconds, and then check "C" seconds for "cooling device". Please send the message "Please."

As described above, the sentence data is divided into sentences on the grammar unit, and the transmission pause is inserted between the divided sentences. Therefore, even if the message is divided into two and the message is transmitted, the voice output is performed. The section 42 can output a natural and easy-to-hear message.

If the "C" seconds are longer than the "T" seconds and the transmission pause is simply inserted between the sentences, the second message is "point the cooling device" as shown in FIG.
In the case where it is further divided into two, such as "Please check.", If it is sent as it is and output as voice, it becomes a message that is very difficult to hear as in the case shown in FIG. ..

Therefore, as shown in FIG.
It is also conceivable to dispose a bunsetsu dividing unit 36 that divides the sentence data into bunsetsu units in grammar in parallel with. in this case,
At the same time that the sentence dividing unit 34 divides the sentence data into grammatical sentence units, the bunsetsu dividing unit 36 grammatically analyzes the sentence data to divide the sentence data into grammatical sentence units, and the divided result is a speech synthesis unit. In step 33, the output time data is sent to the wireless transmitter 35.

By the division processing of the phrase division unit 36, FIG.
As shown in, the above message is further divided into two messages: "E" seconds (E <T) "cooler" and "F" seconds (F <T) "Please check." The wireless transmitter 35 sends a message "Cooling device" for "E" seconds, pauses the sending for "S" seconds, and then sends a message "Check" for "F" seconds. To do so. In this way, the sentence data is divided into bunsetsus on the grammar unit, and by inserting a transmission pause between the bunsetsus, the voice output unit
With 22 you can output a natural and easy-to-hear message.

In the second embodiment, the waveform synthesizing method is adopted as the speech rule synthesizing method in the speech synthesizing section 33, but the present invention is not limited to this, and the formant method, cepstrum, etc. may be used. Of course, other methods may be used.

[0035]

As described above in detail, according to the present invention, in a voice transmitting device for reproducing and transmitting voice recorded by using a wireless device which cannot continuously transmit for a certain time or longer,
A grammatical unit, such as a sentence, that does not exceed a certain length
Since the audio data is divided into a plurality of breaks and recorded in advance and recorded, and the recorded audio data is reproduced and transmitted for each division unit, the reproduced voice can be transmitted. It is possible to provide a voice transmitting device capable of dividing the voice at a natural and easily audible position and dividing and transmitting the voice when it is expected to be paused on the way.

Further, according to the present invention, in a voice transmitting device for transmitting voice data obtained by performing voice rule synthesis from a series of character data using a wireless device which cannot continuously transmit for a certain time or longer, The output time after speech rule synthesis is calculated, and according to the calculated output time, the character data is divided into predetermined grammatical units such as sentences and clauses so as not to exceed a certain time length that allows continuous transmission. Since it is divided into a plurality of pieces, and the divided character data is composed by voice rule and converted into voice data and then wirelessly transmitted every time the above is divided, it is expected that the transmission of the rule-combined voice will be stopped halfway. In this case, it is possible to provide a voice transmitting device capable of automatically dividing the voice at a natural and easily audible position and dividing and transmitting the voice.

[Brief description of drawings]

FIG. 1 is a block diagram showing a circuit configuration according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing an extension example of the circuit of FIG.

FIG. 3 is a diagram for explaining the operation of the embodiment of the present invention.

FIG. 4 is a diagram for explaining the operation of the embodiment of the present invention.

FIG. 5 is a diagram for explaining the operation of the embodiment of the present invention.

FIG. 6 is a diagram for explaining the operation of the embodiment of the present invention.

FIG. 7 is a diagram for explaining the operation of the embodiment of the present invention.

FIG. 8 is a block diagram showing a circuit configuration according to a second embodiment of the present invention.

9 is a block diagram showing an extension example of the circuit of FIG.

[Explanation of symbols]

10, 30 ... Transmitter side, 20, 40 ... Receiver side, 11a, 11b, 11
n ... recording section, 12 ... audio reproducing section, 13 ... wireless transmitting section, 21, 41
... wireless receiving section, 22, 42 ... voice output section, 31 ... character data input section, 32 ... output time calculating section, 33 ... voice synthesizing section, 34 ... sentence dividing section, 35 ... wireless transmitting section, 36 ... phrase dividing section ..

Claims (6)

[Claims] 1. A voice transmitting device for reproducing and transmitting voice data recorded by using a wireless device which is not capable of continuous transmission for a fixed time or longer, in terms of grammar so as not to exceed the fixed time length capable of continuous transmission. A recording means for dividing the audio data into a plurality of pieces by dividing the audio data into a plurality of divisions, a reproducing means for reproducing the audio data recorded in the recording means, and the audio data obtained by the reproducing means. A voice transmitting device, comprising: a transmitting unit that wirelessly transmits each time the signal is transmitted. 2. The recording means divides the voice data into a plurality of divisions by dividing the voice data at sentence breaks.
The voice transmission device described. 3. The voice transmitting apparatus according to claim 1, wherein the recording means records the voice data by dividing the voice data at a break of a phrase and dividing the voice data into a plurality of pieces. 4. A voice transmitting device for transmitting voice data obtained by performing voice rule synthesis from a series of character data, using a wireless device which is not capable of continuous transmission for a certain period of time or more. And an output time calculating means for calculating the output time of the character, and the character is divided by a grammatical unit break so as not to exceed the constant time length for continuous transmission according to the output time obtained by the output time calculating means. Dividing means for dividing the data into a plurality of pieces, voice rule synthesizing means for synthesizing the voice data of the character data divided by the dividing means, and voice data obtained by the voice rule synthesizing means are wirelessly transmitted for each of the divided portions. A voice transmitting apparatus comprising: a transmitting unit. 5. The voice transmitting apparatus according to claim 4, wherein said dividing means divides the character data into a plurality of pieces by dividing the character data at sentence breaks. 6. The voice transmitting apparatus according to claim 4, wherein the dividing unit divides the character data into a plurality of pieces by dividing the character data at a break of a phrase.