JP2000194389A - Information processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce erroneous reading without changing the dictionary data used for read-aloud when an optional sentence is read aloud with voice. SOLUTION: When the sentence inputted from an input part 1 is converted to the voice data by a language processing part 2, when all words aren't registered in a voice conversion dictionary data part 3, a dictionary data part 4 used in KANA-KANJI conversion are retrieved inversely from KANJI to KANA, and relevant reading is obtained. Then, the relevant reading is converted to the voice data to be outputted from an output part 5.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to linguistic processing when an arbitrary sentence is read aloud.

[0002]

2. Description of the Related Art Conventionally, when an arbitrary sentence is read aloud, a speech conversion dictionary that converts word notation into reading and converts the reading into speech data is used, but the corresponding word is not registered. In this case, it was necessary to change dictionary data such as adding a new word to the dictionary or changing the reading priority.

[0003]

SUMMARY OF THE INVENTION An object of the present invention is to solve the above-mentioned conventional problems, and an object of the present invention is to correctly convert an arbitrary sentence into speech data without changing speech conversion dictionary data.

[0004]

SUMMARY OF THE INVENTION In order to solve this problem, the present invention provides a method for reversing the search of kanji from kana-kanji conversion data such as a kana-kanji conversion dictionary, a user dictionary, and a learning dictionary. By inputting a correct reading by the operator, reading of words not registered in the voice conversion dictionary data can be obtained.

[0005] This makes it possible to correctly convert an arbitrary sentence into voice data without changing the voice conversion dictionary data.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS According to the first aspect of the present invention, a dictionary data section for storing readings and notations of words and voice data corresponding to the words, and inputting characters. An information processing apparatus comprising: an input unit; a voice conversion dictionary data unit configured to convert a word input by the input unit into voice data; and an output unit configured to reproduce voice data. Has the effect of reading aloud by voice.

According to the invention of claim 2, in the dictionary data section, in addition to the dictionary data (speech conversion dictionary) for converting words into speech data, dictionary data (e.g., Kana-Kanji conversion dictionary), dictionary data specified by the operator for reading and writing (user dictionary), dictionary data learned by kana-kanji conversion (learning dictionary), etc.
2. The information processing apparatus according to claim 1, further comprising dictionary data used in kana-kanji conversion, wherein words not registered in the voice conversion dictionary are converted from kanji to kana-kanji dictionary data. Has the effect of retrieving and converting it to voice data.

According to a third aspect of the present invention, there is provided a range designating section for designating a range of input characters, a candidate display section for displaying reading candidates of the character string designated for the range, and selecting the displayed reading. The information processing apparatus according to claim 2, further comprising: a selecting unit that performs the reading and a dictionary registering unit that registers the selected reading. By newly selecting the reading of the character string, there is an effect that the reading can be correctly performed.

The invention according to a fourth aspect is characterized in that the invention has a reading input unit which allows an operator to input a reading, and a dictionary registration unit which registers the input reading.
In the information processing device described in the above, if there is no target reading in the displayed reading, the operator can input any reading, even if the word is not registered in the dictionary data, It has the effect of enabling correct reading.

An embodiment of the present invention will be described below with reference to the drawings. (Embodiment 1) FIG. 1 is a configuration diagram of an information processing apparatus according to a first embodiment of the present invention.

1 is an input unit, 2 is a language processing unit, 3 is a voice conversion dictionary data unit, 4 is a dictionary data unit, and 5 is an output unit.

The input unit 1 inputs an arbitrary sentence, and the language processing unit 2 performs a morphological analysis of the input sentence, converts notation into reading, and then converts it into voice data. The voice conversion dictionary data unit 3 stores voice conversion dictionary data, the dictionary data unit 4 stores dictionary data used for kana-kanji conversion such as a kana-kanji conversion dictionary, a user dictionary, and a learning dictionary. And output the output audio data.

The operation of the first embodiment will be described with reference to FIG. It is assumed that the sentence "It is fine weather" is input from the input unit 1 (step 41). The morphological analysis is performed by the language processing unit 2 to divide the word into words "sunny" and "is" (step 42). BUNK the number of words split
Substituting in ATUSUU and putting an initial value 1 in N (Step 4
3). BUNKATUSUU is compared with N, and if it is equal to or less than BUNKATUSUU, the process is continued (step 44). The first word "sunny"
Is searched for in the voice conversion dictionary data section 3 (step 45). If not registered, the dictionary data section 4 is searched from the kanji in the reverse direction (step 4).
6). If a matching word is found, the word is read and converted to "Seiten" (step 47). Next, the audio data is converted into audio data for reproduction (step 48), and the audio data is copied to a buffer (step 49). This is repeated until the last word (step 410), and the output unit 5 outputs the voice data "Sesen" which is copied to the buffer (step 411).

Next, a second embodiment of the present invention will be described with reference to the drawings. (Embodiment 2) Embodiment 2 is based on misreading, etc.
It is an example for explaining the processing when changing the reading.

FIG. 2 is a configuration diagram of an information processing apparatus according to a second embodiment of the present invention. 1 to 5 are the same as those in FIG. Reference numeral 6 denotes a range designation unit, 7 denotes a candidate display unit, 8 denotes a selection unit, and 9 denotes a dictionary registration unit.

A range designation unit 6 designates a range of a character string whose reading is to be changed, a candidate display unit 7 displays candidate readings of the designated character string, and a selection unit 8 selects a displayed reading candidate. Select The dictionary registration unit 9 registers the selected reading in the dictionary data unit 4.

The operation of the second embodiment will be described below with reference to FIG. It is assumed that the text input from the input unit 1 is “Autumn cherry blossom season” and that the output unit 5 outputs “Akizakura Kisetsu” by the operation in the first embodiment. If the user wants to change this to output "Kimasu no Kisetsu", the range specifying unit 6 first specifies the range of "Akisakura" (step 51). The language processing unit 2 obtains the reading candidates “Akizakura” and “Kosumosu” registered in the voice conversion dictionary data unit 3 or the dictionary data unit 4 (step 52) and displays them on the candidate display unit 7 (step 52). 53). When the operator selects "Kumosumo" with the selection unit 8 (step 54), the language processing unit 2 converts the data into audio data to be reproduced (step 5).
5) Change the audio data "Akizakura" copied to the buffer to "Kumosumo" (step 56). The audio data "Cosmos no Kisetsu" is output from the output unit 5 as audio data copied to the buffer (step 5).
7). Further, the dictionary registration unit 9 registers the selected reading “kosumosu” in the dictionary data unit 4 (step 5).
8). The registration may be a learning dictionary or a user dictionary.

Thus, when "Autumn cherry blossom season" is input again from the input unit 1 in the subsequent input, the operation in the first embodiment causes the output unit 5 to output "Kosamo no kisetsu". Become so.

(Embodiment 3) In Embodiment 3, when the reading is changed due to misreading or the like, the processing in the case where there is no target reading in the reading candidates displayed by the operation in Embodiment 2 is described. This is an example to be described.

FIG. 3 is a configuration diagram of an information processing apparatus according to a third embodiment of the present invention. 1 to 7 and 9 are FIGS. 1 and 2
Therefore, the description is omitted. Reference numeral 10 denotes a reading input unit.

The reading input unit 10 inputs reading of a character string specified by the range specifying unit 6. The operation in the third embodiment will be described below with reference to FIG.

The sentence input from the input unit 1 is “Shinonome is” and the output unit 5
Output "" If it is desired to change this to be output as "Shinosame", it is assumed that "Shinosame" is not displayed on the candidate display section 7 due to the operation in the second embodiment (step 53).
In this case, the operator uses the reading input unit 10 to perform “Shinonome”.
(Step 61), the language processing unit 2
It is converted into audio data for reproduction (step 62),
The audio data that was copied to the buffer
Is changed to "Shinome" (step 63). Output unit 5
Then, the voice data "copyright" copied to the buffer is output (step 64). Further, the dictionary registration unit 9 registers the input “Shinonome” reading in the dictionary data unit 4 (Step 65). The registration may be a learning dictionary or a user dictionary.

As a result, when "Shinonome is" is input again from the input unit 1 in the subsequent input, the operation in the first embodiment causes the output unit 5 to output "Shinomenome".

[0024]

As described above, according to the present invention, even if a word that is not registered in the voice conversion dictionary data for converting a notation used for reading an arbitrary sentence into reading is used for kana-kanji conversion. By searching the dictionary data in reverse from kanji, or by inputting the desired reading, the operator does not increase the number of words in the voice conversion dictionary data or change the dictionary contents. This has the advantageous effect that sentences can be read out correctly.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an information processing apparatus according to a first embodiment of the present invention.

FIG. 2 is a configuration diagram of an information processing apparatus according to a second embodiment of the present invention;

FIG. 3 is a configuration diagram of an information processing apparatus according to a third embodiment of the present invention.

FIG. 4 is a flowchart showing an operation according to the first embodiment of the present invention;

FIG. 5 is a flowchart showing an operation according to the second embodiment of the present invention;

FIG. 6 is a flowchart showing an operation according to the third embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input part 2 Language processing part 3 Voice conversion dictionary data part 4 Dictionary data part 5 Output part 6 Range specification part 7 Candidate display part 8 Selection part 9 Dictionary registration part 10 Reading input part

Claims (4)

[Claims] 1. A dictionary data section for storing word readings and notations, and for storing voice data corresponding to the words, an input section for inputting characters, and voice data for inputting the words input by the input section. An information processing apparatus comprising: a voice conversion dictionary data section for converting voice data; and an output section for reproducing voice data. 2. The dictionary data section includes dictionary data (voice conversion dictionary) for converting words into voice data, dictionary data (kana-kanji conversion dictionary) previously mounted on the device for kana-kanji conversion, and operation data. 2. The information according to claim 1, comprising dictionary data used in kana-kanji conversion, such as dictionary data (user dictionary) in which a user specifies reading and writing, and dictionary data learned by kana-kanji conversion (learning dictionary). 3. Processing equipment. 3. A range designating unit for designating a range of an input character, a candidate display unit for displaying a candidate for reading a character string whose range is designated, a selecting unit for selecting a displayed reading, The information processing apparatus according to claim 2, further comprising a dictionary registration unit that registers readings. 4. The information processing apparatus according to claim 3, further comprising: a reading input unit that allows an operator to input readings; and a dictionary registration unit that registers the input readings.