JP2647913B2 - Text-to-speech device - Google Patents

Description

The present invention relates to a text-to-speech apparatus that handles documents such as a personal computer and a word processor, and particularly reads out texts with synthesized speech.

(B) Conventional technology A text-to-speech apparatus having a function of outputting a synthesized voice for reading a program or text of a personal computer has been proposed (Japanese Patent Publication No. 62-8813). In recent years, such a text-to-speech apparatus has been applied to a text review system that can check a text by reading the text or a text-to-speech system for blind people.

(C) Problems to be Solved by the Invention In a text-to-speech reading device such as a text review device, a series of input texts are read from the beginning or a specified position, and are read until a stop key is pressed or a specified position. Sentences are read aloud sequentially. However, if the user is unable to hear the sentence aloud or finds it difficult to listen to the same part again, it is necessary to listen again from the beginning or specify the utterance part again, Operation was troublesome.

(D) Means for Solving the Problems The present invention relates to a text-to-speech apparatus using a speech synthesis circuit capable of reading an arbitrary text, and a return text indicating how much to return from the current text for re-reading. The number is set by the setting means, and the operation of the repeat switch for rereading the already read sentence or the reception of a repeat signal from the outside interrupts the utterance of the sentence being read out, and the It is characterized in that re-reading is started from the beginning of the sentence returned by the number of returned sentences.

(E) Function According to the text-to-speech apparatus according to the present invention, the above-described repeat function has the above-described repeat function. It is possible to instantly check the parts that are difficult to take.

(F) Embodiment FIG. 1 is a configuration diagram of an embodiment of a text-to-speech apparatus for realizing the present invention.

The input sentence is divided into sentences by a sentence boundary discriminating circuit (1) and stored in a sentence buffer memory (2).
FIG. 2 shows an example of a sentence stored in the sentence buffer memory. In the example, the period "." Is determined as a text boundary. The text buffer memory (2) is provided with an address pointer (3) to select a text. The utterance switch (4) is a switch for instructing to perform speech synthesis. When the utterance switch (4) is turned on, the clock (6) of the clock generation circuit (5) is sent to the address pointer (3). . Since the address pointer (3) is previously set to 0 by the initialization switch (7), 1 is set by the clock (6), and the text with the text number 1 shown in FIG. 2 is selected. On the other hand, the clock (6) is sent as an utterance instruction signal (9) to the rule synthesizing circuit (8) via the OR circuit. The rule synthesizing circuit (8) reads out a sentence from the sentence buffer memory (2), converts the sentence into voice, and utters it. During the rule synthesis, a signal (10) is output. Due to the presence of the signal (10), the input of the clock (6) generated by the clock generation circuit (5) to the address pointer (3) is prohibited and the address is output. The contents of the pointer (3) do not change. Also, no utterance instruction signal (9) is generated. Rule composition is completed, ▲ ▼
When the signal (10) is no longer output, the clock (6) generated from the clock generation circuit (5) is sent to the address pointer (3), and the content of the address pointer (3) is 2
The sentence of the sentence number 2 shown in FIG. 2 is selected, and the utterance instruction signal (9) is sent to the rule synthesizing circuit (8). Then, the rule composition of the sentence No. 2 is performed.

Next, the operation when the repeat switch (11) is pressed will be described. For example, if the repeat switch (11) is pressed while the nth sentence is being uttered, the utterance interruption signal (12)
Is sent to the rule synthesizing circuit (8), the rule synthesizing process is interrupted, and the utterance is also interrupted. On the other hand, the utterance instruction signal (9) also
It is sent to the rule synthesizing circuit (8) via the OR circuit. The rule synthesizing circuit (8) reads out a sentence from the sentence buffer memory (2) in accordance with the utterance instruction signal (9) and performs the rule synthesizing process, but is generated by the clock generating circuit (5) by pressing the repeat switch (11). Since the input of the clock (6) to the address pointer (3) is prohibited and the content of the address pointer (3) does not change, the n-th sentence is reselected and the n-th sentence is uttered again from the beginning. Is done.

The embodiment apparatus described above is provided with the repeat switch (1
By pressing 1), the n-th sentence that had been read aloud up to that point is read again from the beginning, but when the repeat switch (11) is pressed, the address pointer (3) is read.
If an appropriate number, for example, 2 is subtracted from the content of
-You can read aloud from the beginning of the second sentence.

Further, the pressing time of the repeat switch (11) is measured by a timer, and the address pointer (3)
Can be configured to determine the magnitude of the subtraction value of.
Further, the number of times the repeat switch (11) is continuously pressed may be subtracted from the value of the address pointer (3). Thus the address pointer (3)
If it is configured to be able to specify the subtraction number of, it becomes possible to trace the start point of re-reading back to an arbitrary preceding sentence.

(G) Effects of the Invention As is clear from the above description, the text-to-speech apparatus of the present invention includes the setting means for setting the number of return texts indicating how much to return from the current text in order to read again. When there is a part that is hard to hear or a part that is difficult to hear, it can be heard again by a simple operation, and there is no need to confirm the text on a display device or move the cursor to specify the utterance position.

In addition, the number of return sentences can be specified by the setting means based on the pressing time of the repeat switch, the number of times of pressing, and the like, so that a text-to-speech apparatus with improved operability can be realized.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of a text-to-speech apparatus of the present invention, and FIG. 2 is a memory diagram showing an example of text stored in a text buffer memory. (2) ... sentence buffer memory, (3) ... address pointer, (4) ... voice switch, (5) ... clock generation circuit, (8) ... rule synthesis circuit, (11) ... repeat switch , AND ... AND circuit, OR ... OR circuit, INV
…… Inverter circuit.

Claims (1)

(57) [Claims] In a text-to-speech apparatus using a speech synthesis circuit capable of reading an arbitrary text, the number of return texts indicating how much to return from the current text for re-reading is set by setting means. By operating the repeat switch for re-reading the text that has already been read or by receiving a repeat signal from the outside, the utterance of the text being read is interrupted, and the beginning of the text returned from the current text by the number of return texts described above. A text-to-speech device that starts reading again from