JPH07129555A - Document reading device - Google Patents

Abstract

(57) [Summary] [Structure] A personal dictionary (notation reading personal dictionary 108) that correlates from notation to reading and a personal dictionary (reading notation personal dictionary) that correlates from reading to notation that has already been registered and created by a document creation device or the like. By including the reading-and-notation correspondence unit 106 that associates the contents of the two dictionaries 109) with each other, the work of registering the document reading device in the personal dictionary is saved. [Effect] (1) It is not necessary to re-register the dictionary information that has already been registered in the "reading personal dictionary". (2) The reading can be generated from the "reading notation personal dictionary" of the document creation device even if it is not in the "notation reading personal dictionary". (3) From the "reading and writing personal dictionary" such as a document creation device,
An "inscription reading personal dictionary" can be automatically created. (4) Since it is read aloud according to the user's favorite reading registered in the "reading and writing personal dictionary", it is easy for the reading device to find a mistake in the writing.

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 device for reading out texts in natural language.

[0002]

2. Description of the Related Art In recent years, a document reading device for reading a sentence in natural language has been researched and developed. The configuration of the conventional document reading device will be described with reference to FIGS. 8 and 9.

FIG. 8 is a diagram showing the structure of a conventional document reading device. In the figure, reference numeral 801 denotes a sentence input unit that receives a natural language sentence that forms a document. A sentence analysis unit 802 analyzes the natural language sentence obtained from the sentence input unit 801, and analyzes the reading and accent for reading aloud. Reference numeral 803 is an acoustic processing unit that generates a speech waveform from the reading and accent of the sentence analyzed by the sentence analysis unit 802. An audio output unit 804 outputs the audio waveform generated by the audio processing unit 803 to the speaker to output audio. A dictionary search unit 805 searches a dictionary used by the sentence analysis unit 802. Reference numeral 806 is a dictionary that is standardly installed in the device. Reference numeral 807 is a personal dictionary in which the user registers information on words that are not in the standard dictionary 806. The dictionary search unit 805 searches the standard dictionary 806 and the personal dictionary 807.

FIG. 9 is a flowchart showing a process of reading a document by the document reading device shown in FIG.

In FIG. 9, when a user's input is received in the input sentence receiving processing step S11, a judgment step S
Pass to 12. In this judgment step S12, in order to search the dictionary information for analyzing the sentence of the input sentence from the standard dictionary 806 and the personal dictionary 807, it is first checked whether or not the information of the word to be searched exists in the personal dictionary 807, and If there is a search word depending on the existence, a personal dictionary search processing step S1
3. If there is no search word, go to standard dictionary search processing step S14.

The personal dictionary retrieval processing step S13 retrieves the personal dictionary 807 and transfers the retrieval result to the sentence analysis processing step S15. Standard dictionary search processing step S14
Searches the standard dictionary 806 and passes the search result to the sentence analysis processing step S15. Sentence analysis processing step S15
Analyzes the sentence from the dictionary search result, analyzes its reading and accent information, and passes the analysis result to the acoustic processing step S16.

The sound processing step S16 generates a voice waveform from the reading and accent information passed from the sentence analysis processing step S15, and a voice output processing step S17.
Pass to. The voice output processing step S17 outputs the passed voice waveform as voice to a speaker or the like.

Then, in step S18, it is determined whether or not the reading of the document is ended, and if it is ended, the reading of the document is ended. If not ended, the process returns to the input receiving process step of step S11.

This repetitive process is the flow of the process of reading the document by the conventional document reading device.

[0010]

However, in an environment where the document reading device is used, it is usual to use a word processor (for example, a Japanese word processor), which is an input device for creating a document to be read, as a document creating device. Is. In this type of document creating device, a personal dictionary that corresponds from reading to notation for words that the user personally uses or words that are not standardly prepared in the document creating device (hereinafter referred to as "reading notation personal dictionary") Are equipped with. Then, the user can register the correspondence from reading to notation according to his or her preference.

However, in such a case, the same information as the information in the dictionary (reading / writing personal dictionary) corresponding to reading to notation of the personal dictionary already registered in the document creation device is used as the correspondence from notation to reading. It is necessary to re-register in a personal dictionary (hereinafter, referred to as “personal expression reading personal dictionary”) that corresponds from reading to reading of the reading device, which requires a complicated operation.

Therefore, in view of the above points, an object of the present invention is to provide a document reading device capable of coping with various flexible reading methods and simplifying the operation therefor.

[0013]

In order to achieve such an object, the present invention is a document reading device for reading a sentence described in natural language, and means for storing a personal reading personal dictionary that corresponds to reading from reading. A user personal dictionary registered in another external device, a means for storing a personalized reading and writing personal dictionary in which the readings and the notations are associated with each other, and a reading and writing correspondence means for associating the dictionary contents of the two personal dictionaries. It is equipped.

[0014]

According to the present invention, a personal dictionary that associates notations to readings (notation reading personal dictionary) and a personal dictionary that has already been registered and created by a document creation device or the like (corresponding to readings and notations personal dictionary). By including the "reading / writing notation correspondence means" for associating the contents of the two dictionaries, the registration work of the document reading device in the personal dictionary can be saved.

[0015]

Embodiments of the present invention will be described in detail below with reference to the drawings.

Embodiment 1 FIG. 1 is a block diagram showing the configuration of a document reading device according to an embodiment of the present invention. In the figure, 101 is
A sentence input unit that receives a natural language sentence in a document. 10
Reference numeral 2 denotes a sentence analysis unit that analyzes the natural language sentence obtained from the sentence input unit 101 and analyzes the reading and accent for reading. Reference numeral 103 is an acoustic processing unit that generates a speech waveform from the reading and accent of the sentence analyzed by the sentence analysis unit 102. Reference numeral 104 is a voice output unit that outputs the voice waveform generated by the acoustic processing unit 103 to a speaker to output voice.

Reference numeral 105 is a dictionary search unit for searching a dictionary used by the sentence analysis unit 102. Reference numeral 107 is a dictionary that is standardly equipped with this device. Reference numeral 108 is a “notation reading personal dictionary” in which the user registers information on the notation and reading of words that are not in the standard dictionary 107 registered in this device. 109
Is a "reading / writing personal dictionary" that associates readings with notations that have already been registered by another document creation device or the like.

Reference numeral 106 designates the above-mentioned "individual reading personal dictionary" 10
8 is a “reading / writing notation correspondence unit” that makes correspondence between 8 and the “reading / writing personal dictionary” 109. The dictionary search unit 10
Reference numeral 5 indicates a “reading notation personal dictionary” 108 and a “reading notation personal dictionary” 109 via the “reading notation correspondence unit” 106.
Then, the standard dictionary 107 provided as standard is searched.

FIG. 2 is a diagram showing a detailed configuration of the "reading / writing notation corresponding part" 106 shown in FIG. In the figure, reference numeral 203 denotes a dictionary information correspondence table showing a correspondence relationship of the dictionary information between the “reading and writing personal dictionary” 109 and the “writing and reading personal dictionary” 108. Reference numeral 202 denotes a dictionary information correspondence unit, which uses the dictionary information correspondence table 203 to convert the “reading notation personal dictionary” 109 into the same format as the “notation reading personal dictionary” 108, and passes the information to the dictionary search unit 201.
The dictionary search unit 201 has two dictionaries, the information converted into the same format as the “notation reading personal dictionary” 108, which is the personal dictionary of the apparatus by the dictionary information corresponding unit 202, and the “notation reading personal dictionary” 108. Is a dictionary search unit for searching for.

FIG. 3 is a flowchart showing a process of reading a document by the document reading device shown in FIGS. 1 and 2.

First, an input sentence is received in the input sentence receiving processing step S1, and is passed to the judgment step S2. In the judgment step S2, it is checked whether or not the dictionary information of the word for analyzing the sentence of the input sentence exists in the "writing reading personal dictionary", and if it exists, the "writing reading personal dictionary" search processing step S
3, the process proceeds to the judgment step S4 if it does not exist.

In the judgment step S4, it is checked using the dictionary information correspondence table 203 and the dictionary information correspondence unit 202 whether or not the information of the search word exists in the "reading and writing personal dictionary" 109. If it exists, the process proceeds to "reading personal dictionary" search processing step S5, and if it does not exist, the standard dictionary search processing step S6 proceeds.

"Notation reading personal dictionary" search processing step S
3 retrieves a "notation reading personal dictionary" that holds the word notation and reading and accent information as shown in FIG. 4, and passes the result to the sentence analysis processing step S7.

"Reading personal dictionary" search processing step S
5 is a dictionary information correspondence table 203, an example of which is shown in FIG.
5, the dictionary information correspondence unit 202 makes a correspondence between reading and notation, searches the “reading notation personal dictionary” 109, an example of which is shown in FIG. 5, by notation, and passes the result to the sentence analysis processing step S7.

In the standard dictionary search processing step S6, the standard dictionary 107 is searched, and the result is sentence analysis processing step S7.
Pass to. The sentence analysis processing step S7 is S3, S5, S6.
Sentences are analyzed from the results of each dictionary search process of
The accent information is analyzed, and the result is passed to the acoustic processing step S8.

In the acoustic processing step S8, a voice waveform is generated from the reading accent information passed from the sentence analysis processing step S7, and the voice waveform is passed to the voice output processing step S9. The voice output processing step S9 outputs the passed voice waveform as voice to a speaker or the like.

Then, in step S10, it is determined whether or not the reading of the document is ended, and if it is ended, the reading of the document is ended. If not ended, the process returns to the input receiving process of step S1.

This repetitive process is the process flow of reading a document by the document reading device.

Second Embodiment In the first embodiment described above, the "reading and writing personal dictionary" 10 is used by using the dictionary information correspondence unit 202 during the repetition of reading the document.
9 was converted and searched, but the present invention is not limited to this, and can be collectively registered in the "notation reading personal dictionary" 108 before repeating reading of the document.

An example of the "reading / writing correspondence unit" 106 in this case will be described with reference to FIG.

The dictionary update unit 401 uses the dictionary information correspondence table 203 and the dictionary information correspondence unit 202 to collectively convert the "reading notation personal dictionary" 109 into the "notation reading personal dictionary" 108 format, "Notation reading personal dictionary" 108
It is a dictionary update unit registered in. As a result, the information of the personal dictionary already registered by the user using the document creation device or the like is automatically registered in the "writing notation personal dictionary" 108, so that the user does not have to newly register again.

With this configuration, the process of reading the document by the document reading device is a process except for the judgment unit S4 and the "reading / writing personal dictionary" search process S5 shown in FIG.

Third Embodiment In the first embodiment described above, an example of a personal dictionary in which the notation of a Japanese document is associated with reading is described, but the present invention is not limited to this, and the same applies to a voice rule synthesizer for a language other than Japanese. With the configuration, it is possible to create the present invention that saves the work of registering a personal dictionary that associates notation with reading.

[0034]

As described above, the contents of the "reading / writing personal dictionary" already registered by the user with another document creating device and the contents of the "writing / reading personal dictionary" of the document reading device itself are associated with each other. By providing "reading and writing correspondence means", the effects listed below can be obtained.

(1) It is not necessary to re-register the dictionary information that has already been registered in the “reading and writing personal dictionary”.

(2) It is possible to generate a reading from the "reading notation personal dictionary" of the document creation device even if it is not in the "notation reading personal dictionary".

(3) An "inscription reading personal dictionary" can be automatically created from the "reading notation personal dictionary" of a document creation device or the like.

(4) Since the reading is carried out according to the user's favorite reading registered in the "reading / writing personal dictionary", it is easy to find an error in the writing by the reading device.

[Brief description of drawings]

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

FIG. 2 is a block diagram showing a reading / writing correspondence unit of the document reading device according to the first embodiment of the present invention.

FIG. 3 is a flowchart showing a processing procedure of the document reading device according to the first embodiment of the present invention.

FIG. 4 is a diagram showing an example of contents of a “notation reading personal dictionary” in the first embodiment of the present invention.

FIG. 5 is a diagram showing an example of contents of a “reading and writing personal dictionary” in the first embodiment of the present invention.

FIG. 6 is a diagram showing an example of contents of a dictionary information correspondence table according to the first embodiment of the present invention.

FIG. 7 is a block diagram showing a reading / writing notation corresponding part of a document reading device according to a second embodiment of the present invention.

FIG. 8 is a block diagram showing a conventional document reading device.

FIG. 9 is a flowchart showing a processing process in a conventional document reading device.

[Explanation of symbols]

 101 sentence input unit 102 sentence analysis unit 103 acoustic processing unit 104 voice output unit 105 dictionary search unit 106 reading and writing correspondence unit 107 standard dictionary 108 writing reading personal dictionary 109 reading and writing personal dictionary

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁶ Identification code Internal reference number FI technical display location 8125-5L G06F 15/38 C

Claims (1)

[Claims] 1. A document reading device for reading a sentence written in a natural language, comprising means for storing a personal reading personal dictionary that associates readings with readings, and a user personal dictionary registered in another external device. A document reading device comprising: a means for storing a reading and writing personal dictionary associated with reading and writing; and a reading and writing correspondence means for associating the dictionary contents of the two personal dictionaries.