JP2002351484A - Voice reading-aloud device and its control method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a voice reading-aloud device which can improve the operability when an operator tries to listen to an intended sentence again and its a control method. SOLUTION: The voice reading-aloud device equipped with a voice output part 50 which outputs a document through voice synthesis, is equipped with a signal input part 70 which receives the input of an instruction to re-listen from the operator and a calling control part 60 which stops the voice output of the voice output part 50 when the signal input part 70 receives the input of the instruction of re-listening from the operator while the voice output part 50 outputs the voice and controls the voice output part 50 to voice the sentence whose voice output is stopped from the beginning of the sentence or the beginning of the sentence before the sentence whose voice output is stopped.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a text-to-speech apparatus for outputting a text sentence by speech synthesis and a control method thereof.

[0002]

2. Description of the Related Art Some conventional text-to-speech devices output texts such as e-mails and newspaper articles by voice synthesis and read out the contents.

[0003] In this type of apparatus, a text-to-speech apparatus starts speech output from the first sentence of a text sentence, or an operator determines a sentence to be output as a text sentence by visual information and designates the sentence. It is common for a device to start outputting speech from a specified sentence of a text sentence.

[0004]

However, when the above-mentioned voice reading apparatus is used in a vehicle, for example, the contents can be understood due to the low quality of voice synthesis, a decrease in thinking ability during driving, and missed hearing. Sometimes, you may want to listen again. In this case, while the vehicle is stopped, it is possible for the operator to determine the sentence to be output by voice based on the visual information and specify the sentence, but this is not possible for safety during traveling.

[0005] In this case, there is a problem that when re-listening, the voice is output from the first sentence of the text sentence.

Accordingly, an object of the present invention is to provide a voice reading apparatus and a control method therefor which can solve the above-mentioned problems of the prior art and can improve the operability when re-listening a sentence intended by an operator. Is to do.

[0007]

According to the first aspect of the present invention,
In a voice reading apparatus including a voice output unit that outputs a sentence by voice synthesis, a signal input unit that receives an input of a re-listening command from an operator, and the signal input unit is configured to output a voice by the voice output unit. When receiving an input of a re-listening command from the operator, the voice output of the voice output means is stopped, and the first sentence of the sentence in which the voice output is stopped, or the first sentence of the sentence prior to the sentence in which the voice output is stopped is referred to as Control means for performing control for outputting sound by the sound output means.

According to a second aspect of the present invention, there is provided a voice reading apparatus provided with a voice output means for outputting a sentence by voice synthesis, wherein a signal input means for receiving an input of a re-listening command from an operator, and the voice output. When the signal input means receives an input of a re-listening command from the operator during the sound output by the means, the sound output of the sound output means is stopped,
Control means for controlling the voice output means to output a voice from the beginning of a sentence in which speech output has been stopped, or the beginning of a sentence prior to the sentence in which speech output has been stopped, according to the form of the re-listening command. It is characterized by the following.

According to a third aspect of the present invention, in the first or second aspect of the present invention, the signal input means is a voice detection means for receiving a re-listening command by voice from an operator. It is assumed that.

According to a fourth aspect of the present invention, in the first or second aspect of the present invention, the signal input means is a tactile sense detecting means for receiving a tactile rehearsal command from the operator. It is assumed that.

According to a fifth aspect of the present invention, in the first or second aspect of the present invention, the signal input means includes a voice detection means for receiving an instruction for re-listening by voice from the operator, And a tactile sense detecting means for receiving an input of a re-listening command by the tactile sense.

The invention according to claim 6 is characterized in that the main body is mounted on a vehicle.

According to a seventh aspect of the present invention, there is provided a control method of a voice reading apparatus for outputting a sentence by voice synthesis, wherein the voice output is stopped when a re-listening command is input from the operator during the voice output. Then, control is performed to output a voice from the beginning of a sentence in which speech output has been stopped or from the beginning of a sentence before the sentence in which speech output has been stopped.

According to an eighth aspect of the present invention, in the control method of the voice reading aloud apparatus for outputting a text by voice by voice synthesis, the voice output is stopped when a re-listening command is input from the operator during the voice output. After that, according to the form of the re-listening command, control is performed such that voice is output from the beginning of the sentence in which speech output has been stopped or from the beginning of a sentence before the sentence in which speech output has been stopped. .

The first aspect of the invention has the following operation.

In the control means, when the signal input means receives a re-listening command from the operator while the voice output means is outputting voice, the voice output of the voice output means is stopped and the voice output is stopped. Since the voice output means controls the voice output from the beginning of the sentence or the beginning of the sentence before the sentence of which the voice output has been stopped, it is possible to improve the operability when the operator listens to the sentence intended again. it can.

The second aspect of the invention has the following operation.

In the control means, when the signal input means receives an input of a re-listening command from the operator while the voice output means outputs the voice, the voice output of the voice output means is stopped and the re-listening command is issued. Depending on the form, the voice output means controls the voice output from the beginning of the sentence in which the voice output is stopped or from the beginning of the sentence before the voice in which the voice output is stopped, so that the operator can listen to the sentence intended by the operator. The operability at the time of repair can be improved.

The third aspect of the invention has the following operation.

Since the signal input means is a voice detection means,
Since a re-listening command can be input by voice, operability when re-listening a sentence intended by the operator can be improved.

The invention described in claim 4 has the following operation.

Since the signal input means is a tactile detection means,
By pressing the tactile detection means, for example, a re-listening command can be input, so that operability when re-listening a sentence intended by the operator can be improved.

The invention described in claim 5 has the following operation.

Since the signal input means is the voice detection means and the tactile detection means, a re-listening command can be input by voice, and a re-listening command can be input by pressing the tactile detection means, for example. The operability when re-listening the intended sentence can be improved.

The invention according to claim 6 has the following operation.

Since the operator can input a re-reading command even while the vehicle is driving, the operability when the operator re-reads a sentence intended during driving can be improved.

The invention according to claim 7 has the following operation.

When a re-listening command is input from the operator during voice output, after the voice output is stopped, the beginning of the sentence where the voice output is stopped or the sentence before the sentence where the voice output is stopped is output. Since the control for outputting a voice from the beginning of the sentence is performed, the operability when the operator re-listens for the intended sentence can be improved.

The invention according to claim 8 has the following operation.

When a re-listening command is input from the operator during voice output, after the voice output is stopped, the beginning of the sentence for which the voice output is stopped, according to the form of the re-listening command,
Alternatively, since the control is performed so that the voice is output from the beginning of the sentence before the sentence in which the voice output is stopped, the operability when the operator re-listens the sentence intended by the operator can be improved.

[0031]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, an embodiment of the present invention will be described.
This will be described with reference to the drawings.

FIG. 1 is a block diagram of a speech reading apparatus according to the present embodiment. 1 is a voice reading device (main body)
The text-to-speech apparatus 1 includes a text sentence storage unit 10, a text sentence analysis unit 20, and an analysis sentence storage unit 3.
0, a calling unit 40, a voice output unit 50, a calling control unit 60, and a signal input unit 70. The voice reading device 1 is mounted on, for example, a vehicle.

The signal input section 70 is provided with a voice detection device 80 composed of, for example, a microphone and a tactile device 90 composed of, for example, a push button switch.

The text document storage unit 10 stores text documents such as e-mails and newspaper articles. For example, "The effect of the developed low pressure caused heavy snow around the Sea of Japan. The snow in each area was 40 cm in the plains of the Tohoku region,
It is 80cm in many places such as mountains.
In addition, it is expected that the peak of the snow will be over and that it will gradually recover. It will be mostly sunny tomorrow. Is stored.

The text sentence analyzer 20 divides the text sentence in the text sentence storage 10 into sentence units. Here, a sentence is terminated by a period or separated by a blank line feed.

In the analysis sentence storage unit 30, sentence groups analyzed by the text sentence analysis unit 20 are sequentially numbered from 1 and stored from the first sentence. For example, the following is stored as follows.

1 Due to the developed low pressure, heavy snowfall occurred mainly on the Sea of Japan side.

2 The snow cover in each area is 40 in the plains of the Tohoku region.
cm, 80cm in many places such as mountainous areas.

3 In addition, it is expected that the peak of the snow will be exceeded and the snow will gradually recover.

4 Tomorrow will be mostly sunny.

The total number is MAX_K, and in this example, M
AX_K = 4.

The calling unit 40 sets K = 1 to K = MAX
Each time the voice output of one sentence ends up to _K, the sentence is called from the analyzed sentence storage unit 30 in numerical order. Here, K is a number corresponding to the sentence that is being output as voice. In the above example, K
= 1 to K = 4 are sequentially called from the analysis sentence storage unit 30 by the calling unit 40.

The voice output unit 50 outputs voice of the sentence called from the calling unit 40 by voice synthesis.

The signal input unit 70 is a device for receiving an input of a re-listening command from the operator. This re-listening command is input from the voice detection device 80 or the tactile device 90. When the operator inputs a re-listening command from the voice detection device 80, the operator inputs the voice command of the re-listening command, for example, by saying “return”. When inputting from the haptic device 90, the input is performed when the operator presses the push button switch which is the haptic device 90. Here, the haptic device 90 may be a touch switch or a joystick, and when the touch switch is used,
The input may be made by touching the touch switch, and when the joystick is used, the input may be made when the lever of the joystick is pulled upward, for example.

The calling control unit 60 receives a re-listening command from the operator to the signal input unit 70 during voice output,
When a signal of a re-listening command from the signal input unit 70 is detected, control is performed to stop the sound output of the sound output unit 50 via the calling unit 40. Next, the call control unit 60
The analysis sentence storage unit 30 stores the sentence in which the sound output is stopped in the calling unit 40 or the sentence with the number preceding the sentence in which the sound output is stopped.
Control to call from. The sentence called by this control is output as a voice by the voice output unit 50 from the beginning of the called sentence.

A specific control method will be described with reference to the flowchart of FIG.

FLAG in FIG. 2 is the number of times a command for re-listening is received by the signal input unit 70 until the output of one sentence is completed, and the initial value of FRAG at the start of the output of one sentence is 0. . Here, a description will be given assuming that the sentence output as speech is “No. 3 of the above example sentence,“ It is expected that the snow peak will be exceeded and will gradually recover from now on ”.

First, it is determined whether the input to the voice detection device 80 or the tactile device 90, which is the signal input unit 70, is a re-listening command (S1). That is, a re-listening command is input by uttering “return” in the microphone that is the voice detection device 80 or a re-listening command is input by pressing a switch in the push-button switch that is the haptic device 90. Is determined. Here, if an utterance other than “return” is input to the voice detection device 80, it is not determined that the instruction is a re-listening.

If the input to the signal input unit 70 is a re-listening command, FLAG = FLAG + 1 is set (S
2), S2 is repeated as many times as the number of times a re-listening command is received. As a result, the FRAG becomes a value corresponding to the number of re-listening instructions.

Next, it is determined that FLAG> 1 (S
3). Here, if FLAG = 1, it means that the re-listening command has been input once, and if FLAG> 1, it means that the re-listening command has been input twice or more.

When FLAG> 1 is not satisfied (that is, when FLAG
= 1), M is used to specify the sentence that is outputting voice.
= K (S4). For example, when a re-listening command is input once while the sentence of the example sentence No. 3 is being output as voice,
M = 3. That is, the sentence No. 3 of the above example sentence is designated. Here, K is the number of the sentence output by the voice output unit 50, and M is a temporary variable.

When FLAG> 1, M = K−1 is set in order to specify the sentence immediately before the sentence output as voice (S
5). For example, when a re-listening command is input twice or more while the sentence of the example sentence No. 3 is being output as voice, M = 2. That is, the sentence of number 2 immediately before the above example sentence is specified.

Incidentally, in S4 and S5, although not shown, for example, M = K-1 is set in S4, and M = K-1 in S5.
If it is set to K-2, the command for re-listening during voice output is 1
If the input is performed twice, the sentence having the number preceding the sentence output by S4 is specified in S4. If the re-listening command is input twice or more during the voice output, the voice is output in S5. The sentence with the number two before the sentence number is specified.

If M> 0 is determined (S6), and if M> 0 is not satisfied, the first sentence is specified because the re-listening command has been input to the signal input unit 70 twice or more in the first sentence. M is set to 1 (S7). For example, if the re-listening command is input twice or more while the sentence of the example sentence No. 1 is being output as speech, M = 0 is set in S5, but M is set in S6.
> 0 is determined and M> 0 is not satisfied, so that M = 1 is set in S7. That is, the sentence number 1 of the above example sentence is specified.

The audio output of the audio output unit 50 is stopped, and K
= M (S8).

After the sentence of the number K to be voice-outputted is called from the analytic sentence storage unit 30 to the calling unit 40, the voice output unit 50 starts voice output from the beginning of the sentence of the number K (S).
9). For example, if M = 3, K is set to K = 3 in S8, and the sentence of the number 3 to be output as voice in S9 is stored in the analytic sentence storage unit 3.
After being called by the calling unit 40 from 0, the sound output unit 50 starts sound output from the beginning of the sentence of the number 3.

Thereafter, the flow returns to S1, and if there is no re-listening command in S1, the voice output of one sentence ends (S10), and the sentence of the next number is output as voice.

Here, if it is desired to re-listen to the sentence with the number two before the sentence number currently being output according to this flowchart, the above-described operation of re-listening for the sentence with the immediately preceding number is repeated twice. Is achieved by

According to the present embodiment, the signal input unit 70 that receives an input of a re-listening command from the operator is the microphone that is the voice detection device 80 that receives the input of a re-listening command by the operator. And a push-button switch, which is a haptic device 90 that receives a tactile input from the operator. Therefore, when the operator issues a re-hearing command while driving the vehicle, the re-hearing command is input. The operation of inputting a re-listening command can be performed by the voice detecting device 80 or the tactile device 90 without moving the line of sight to the screen in order to perform the operation described above, and the operability when re-listening the sentence intended by the operator is improved. be able to.

Even when the user wants to hear the sentence before the sentence being output as a voice again, the user does not need to use another voice command as a replay command, for example, a different voice command such as "next". Since the operation can be performed only by the voice command of “return”, the operation can be performed without any uncomfortable feeling for the operator, so that the operability when re-listening the sentence intended by the operator can be improved.

Further, when the operation is performed by the push button switch as the haptic device 90, the intended sentence can be heard again by a simple operation such as pressing the push button switch. The operability at the time of repair can be improved.

As described above, the present invention has been described based on one embodiment, but the present invention is not limited to this.

In the above-described embodiment, when a re-listening command is input to the signal input unit 70 by the operator, the voice output is stopped, and the voice output is performed according to the form of the re-listening command (the number of inputs). Has been described in the case where the speech output unit 50 outputs the speech from the beginning of the sentence or the beginning of the sentence before the sentence in which the speech output is stopped, but the re-listening command is input to the signal input unit 70. In this case, the voice output is stopped, and the sentence in which the voice output is stopped or the sentence prior to the sentence in which the voice output is stopped is immediately called by the calling unit 40 without judging the number of times of inputting the re-listening command. The voice output unit 50 may output voice from the beginning of the sentence.

Further, in the above embodiment, when one voice command, for example, the same voice command of only “return” is input from the voice detection device 80, according to the number of inputs of the re-listening command, A case has been described in which the speech output unit 50 outputs a speech from the beginning of a sentence in which speech output has been stopped, or the beginning of a sentence before the sentence in which speech output has been stopped, but the present invention is not limited to this. When a different voice command is input from the voice detection device 80 with the command of
For example, when the first voice command is input, the voice is output from the beginning of the sentence where the voice output is stopped, and when the second voice command is input, the voice is output from the beginning of the sentence before the sentence where the voice output is stopped. The output unit 50 may output audio. In this way, for example, when the operator wants to hear the previous sentence again, the operator may input a corresponding voice command from the voice detection device 80.

[0065]

According to the present invention, it is possible to improve the operability when re-listening a sentence intended by the operator.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a voice reading device according to the present invention.

FIG. 2 is a flowchart of a control method of the voice reading device shown in FIG.

[Explanation of symbols]

 Reference Signs List 1 voice reading device 50 voice output unit (voice output unit) 60 call control unit (control unit) 70 signal input unit (signal input unit) 80 voice detection device (voice detection unit) 90 tactile device (tactile detection unit)

 ────────────────────────────────────────────────── ─── Continued on the front page (72) Inventor Masaki Tosu 5-35-2 Hakusan, Bunkyo-ku, Tokyo F-term (reference) in Clarion Co., Ltd. 5D015 KK04 5D045 AA02

Claims (8)

[Claims] 1. A speech reading apparatus comprising a speech output device for outputting a sentence by speech synthesis, comprising: a signal input device for receiving an input of a re-listening command from an operator; When the signal input means receives a re-listening command from the operator, the sound output of the sound output means is stopped, and the beginning of the sentence where the sound output is stopped, or the sentence before the sentence where the sound output is stopped. And a control unit for controlling the voice output unit to output a voice from the beginning of a sentence. 2. A speech reading apparatus comprising a speech output means for outputting a sentence by speech synthesis, comprising: a signal input means for receiving an input of a re-listening command from an operator; When the signal input unit receives an input of a re-listening command from the operator, the voice output of the voice output unit is stopped, and according to the form of the re-listening command, the beginning of a sentence in which the voice output is stopped, Or a control unit that controls the voice output unit to output a voice from the beginning of a sentence before the sentence in which the voice output is stopped. 3. The voice reading device according to claim 1, wherein the signal input device is a voice detection device that receives an instruction to re-listen by voice from an operator. 4. The voice reading device according to claim 1, wherein the signal input unit is a tactile detection unit that receives a tactile re-listening command from an operator. 5. The signal input means includes: a voice detection means for receiving an instruction for re-listening by voice from an operator;
3. A tactile sense detecting means for receiving an input of an instruction to rehearse by a tactile sense from an operator.
The voice reading aloud device according to 1. 6. The voice reading device according to claim 1, wherein the main body is mounted on a vehicle. 7. A control method of a voice reading device for outputting a text by voice synthesis by voice synthesis, wherein when a re-listening command is input from an operator during voice output, the voice output is stopped and then the voice output is stopped. A method for controlling a speech reading apparatus, wherein the speech output is performed from the beginning of a stopped sentence or the beginning of a sentence preceding a sentence in which voice output is stopped. 8. A control method for a speech reading apparatus for outputting a sentence by speech by speech synthesis, wherein when a re-listening command is input from the operator during the speech output, the speech output is stopped and then the re-listening is performed. A method for controlling a speech reading apparatus, comprising: performing control to output a voice from a sentence beginning of a sentence in which speech output is stopped or a sentence before a sentence in which speech output is stopped in accordance with a form of a command.