JPH07244491A - Conversion system for text sound - Google Patents

Abstract

PURPOSE:To provide a system which is excellent in operation convenience for a user and can extract unknown words and can process it afterward, in a conversion system for text sound which is provided with a text analyzing section, a parameter generating section for synthesizing sound, sound synthesizing section and a sound outputting section, and obtains a sound output from text data. CONSTITUTION:This system is provided with an unknown words extracting means 23 which extracts successively unknown words included till a termination in an input text data at a step of morpheme analysis processing, a transmitting means 25 for unknown words which transmits unknown words extracted already when processing reaches the termination of the input text data, an information inputting means 27 to which a user inputs information required for registering transmitted unknown words to a word dictionary, and an information registering means 29 which registers unknown words and inputted information relating to unknown words to the word dictionary.

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 conversion system for converting text data into a voice signal and outputting the voice signal.

[0002]

2. Description of the Related Art A system for adding an accent or intonation to input text data and outputting a voice signal corresponding to the text data is known as a text-to-speech conversion system. Known forms of implementing this type of system include: a form composed of a text analysis device and a rule synthesis device; a form composed of a host device (sometimes called a text generation device) and a text-to-speech conversion device. Has been. Both forms of the system, a text analysis unit that generates a phoneme, a prosodic symbol string by performing a morpheme analysis process that cuts text data into morphemes such as words with reference to a word dictionary, and a phonological / prosodic assignment process that follows, A synthesizing parameter generation unit that converts a phoneme and a prosodic symbol string into a voice synthesis parameter using a voice unit storage unit, a voice synthesis unit that converts a voice synthesis parameter into voice waveform data, and an analog voice signal from the voice waveform data. It is equipped with a voice output unit that converts and outputs.

[0003]

However, the conventional text-to-speech converter has the following problems.

(1) If there is a word in the text data that is not registered in the word dictionary, a so-called unknown word, the correct reading of that word cannot be obtained.

(2) The user does not know which word in the text data is treated as an unknown word. This is because the user can determine whether or not there is an unknown word in the text data only by knowing what word is registered in the word dictionary. This is because it is practically difficult for the user to remember a large number of words in the dictionary. After all, the problems of (1) and (2) above were detected when the voice was output.

(3) Further, even if unknown words in the text data to be inputted can be determined, the user must enter the unknown word registration mode or the like in order to register the unknown words in the word dictionary. Since it needs to be changed, it is not easy to use.

(4) Further, when the same unknown word appears a plurality of times in a large number of text data to be input, the problems (1) and (2) at least occur each time.

Further, there are the following problems.

(A) When a plurality of users take turns using the text-to-speech conversion system, the user dictionary in the system corresponds only to a single user, so that the old user changes the user dictionary every time the user changes. The contents have to be backed up to an external medium (eg floppy disk) and the new user has to download the user dictionary from the external medium into the user dictionary area of the system, which is time consuming and tedious.

(B) If the user updates the user dictionary while using the system and forgets to back up the contents of the user dictionary to an external medium at the end of using the system, the updated user dictionary of the old user on the system is , It will not be used in the future because it will be erased by the new user.

[0011]

[Means for Solving the Problems] Therefore, the above (1) to (4)
In order to solve the above problem, in the first invention of this application, a phoneme and a prosodic symbol string are generated from text data by a process including a morphological analysis process performed by referring to a word dictionary and a subsequent phoneme / prosodic process. A text analysis unit, a generated phoneme, a synthesis parameter generation unit that converts the prosodic symbol string into a voice synthesis parameter using the voice unit storage unit, and a voice that converts the generated voice synthesis parameter into voice waveform data. A text-to-speech conversion system that includes a synthesis section and a speech output section that converts the generated speech waveform data into an analog speech signal and outputs the analog speech signal.In addition, (i) unknown words included up to the end of the input text data Unknown word extraction means that sequentially extracts at the stage of morphological analysis processing, and (ii) the unknown words that have been extracted so far when the end of the input text data is reached. And the unknown word transmission means to tell the The, (i
ii) Information input means for the user to input the necessary information for registering the transmitted unknown words in the word dictionary, and (iv) registering these unknown words and the information input for the unknown words in the word dictionary. Claims a text-to-speech conversion system with a means for registering information.

In order to solve the above problem (a), in the second invention of this application, a process including a morphological analysis process performed by referring to a word dictionary and a subsequent phonological / prosodic process process is performed. Text analysis means for generating phonemes and prosodic symbol strings from text data, synthesis parameter generating means for converting the generated phoneme and prosodic symbol strings into speech synthesis parameters using a speech segment storage unit, generated speech synthesis A text-to-speech conversion system comprising a voice synthesizing unit for converting a voice parameter into voice waveform data, and a voice output unit for converting the generated voice waveform data into an analog voice signal and outputting the analog voice signal. A standard dictionary in which predetermined words are registered, a user dictionary updated and modified by the user according to the use of the system, and words from these standard dictionary and user dictionary In a text-to-speech conversion system having a dictionary table for searching
Further, (I) means for securing an area for each of a plurality of users in at least the user dictionary of the word dictionary, and (II)
A password input means for inputting a password for identifying the plurality of users, and (III) selectively selecting an area corresponding to the password input from the password input means out of the areas for each of the plurality of users in the word dictionary. A text-to-speech conversion system is provided, which comprises a dictionary selection means for making the above effective and (IV) means for modifying the dictionary table.

In order to solve the above problem (b), in the third invention of this application, text processing is performed by a process including a morphological analysis process performed by referring to a word dictionary and a subsequent phonological / prosodic process. Text analysis means for generating phoneme and prosody symbol string from data, synthesis parameter generation means for converting the generated phoneme and prosody symbol string into speech synthesis parameters using the speech segment storage section, and generated speech synthesis A text-to-speech conversion system comprising a voice synthesis unit for converting parameters into voice waveform data, and a voice output unit for converting the generated voice waveform data into an analog voice signal and outputting the analog voice signal, further comprising (A) a word dictionary There is a user dictionary change detection means for detecting whether or not the information of the user dictionary of is changed during the text-to-speech conversion work, and (B) the user dictionary is changed. If claiming the text-to-speech conversion system comprising a user dictionary change transmission means to convey the fact to the user.

In order to solve the problems (1) to (4), (a) and (b), respectively, in the fourth invention, the first to
Claim a text-to-speech conversion system that includes all of the components of the third invention.

[0015]

According to the first aspect of the invention, when an unknown word appears in the text data, all unknown words (in the case of one or a plurality of them) appearing in the text data in units of input text data. Notify users automatically and collectively. Further, since the unknown word is extracted at the stage when the morphological analysis result is obtained, the unknown word extraction processing time can be shortened as compared with the case where the unknown word is extracted after the phoneme and prosody addition processing. Further, since the necessary information about the unknown word input by the user from the information input means is registered in the word dictionary by the information registration means, the same character or character string as the unknown word is included in another text data to be input thereafter. Even if appears, this character or character string is no longer considered an unknown word.
Further, in the case of the configuration in which the unknown word buffer is used to check the unknown word, it is possible to prevent the same unknown word from being extracted in duplicate from the input text data.

Further, according to the configuration of the second invention, areas for each of a plurality of users can be secured in the word dictionary, and those areas can be selectively and effectively used by the password.
The dictionary table required for searching the user dictionary and the standard dictionary in the selected user area can be generated by the dictionary table correction means.

Further, according to the configuration of the third invention, when the information of the user dictionary in the word dictionary is changed during the text-to-speech conversion work, this point is transmitted to the user, so that the user has a backup of the user dictionary. Can be encouraged.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Each embodiment of the first to fourth inventions of this application will be described below with reference to the drawings. It should be noted that the drawings used for the description are only schematically shown so that the present invention can be understood. Further, in each drawing, the same components may be denoted by the same reference numerals, and the duplicate description thereof may be omitted. As a form for realizing the text-to-speech conversion system: a form composed of a text analysis device and a rule synthesizing device, a form composed of a higher-level device (sometimes referred to as a text generation device) and a text-speech conversion device. It is already known that is known. The inventions of this application are applicable to both the system of the above-mentioned form and the system of form, but in each of the following embodiments, an example in which each invention of this application is applied to the text-to-speech conversion system of the form is described. To do.

1. Example of first invention 1-1. First Embodiment of First Invention FIG. 1 is a block diagram showing a configuration of a text-to-speech conversion system of a first embodiment of the first invention, and FIG. 2 is a system of the first embodiment, a host device 100 and a text-speech conversion. Device 20
It is the block diagram which showed the example of the correspondence with the system of the form which consists of 0. The host device 100 can be composed of, for example, a word processor or a personal computer. This is because these have an input means such as a keyboard, a display means such as a display, and a communication function, and are therefore suitable as means for configuring a host device.

As shown in FIG. 1, the text-to-speech conversion system of the first embodiment of the first invention includes a morphological analysis process performed by referring to the word dictionary 11 and a subsequent phonological / prosodic process process. A text analysis unit 13 that generates a phoneme and a prosodic symbol string from text data by processing, and a synthesis parameter generation unit 1 that converts the generated phoneme and a prosodic symbol string into a speech synthesis parameter using a speech unit storage unit 15.
7, a voice synthesizing unit 19 for converting the generated voice synthesizing parameter into voice waveform data, and a voice output unit 21 for converting the generated voice waveform data into an analog voice signal and outputting the analog voice signal. Further, this system further comprises an unknown word extracting means 23 for sequentially extracting unknown words included in the input text data up to the end in the stage of morphological analysis processing.
The unknown word transmitting means 25 and the unknown word transmitting means 25 which collectively convey the unknown words extracted by the unknown word extracting means 23 to the user when the end of the input text data is reached.
The information input means 27 for the user to input the information necessary for registering the unknown word transmitted from the word dictionary, and the information for registering the unknown word and the information input for the unknown word in the word dictionary 11. It has a registration means 29. Note that reference numeral 23a in FIG. 1 denotes a storage unit (hereinafter referred to as an unknown word buffer) for storing the extracted unknown word.

Here, the text analysis unit 13, the synthesis parameter generation unit 17, the unknown word extraction unit 23, the unknown word buffer 23a, a part of the unknown word transmission unit 25 and the information registration unit 2 are described.
In this embodiment, 9 is composed of a control unit 201 (for example, CPU, register, etc.) provided in the text-to-speech conversion device 200 (details will be described later). Further, the word dictionary 11, the voice unit storage unit 15, the voice synthesis unit 19, and the voice output unit 21 can be configured by known ones. Hereinafter, these constituent components will be sequentially described.

First, as shown in FIG. 3, the word dictionary 11 is composed of, for example, a standard dictionary 11a composed of a ROM in which predetermined words are registered, and a RAM composed of a standard dictionary 11a, for example, according to the use of the system. A user dictionary 11b capable of adding and deleting (updating, correcting) words by
A dictionary table 11c configured by AM for searching words from the standard dictionary and the user dictionary, and a standard dictionary table 11 used for searching words only from the standard dictionary 11a and used as a reference when modifying the dictionary table 11c.
with d. Since the standard dictionary 11a has morphemes at the 1-character level, the input composed of the character character code is a morpheme at the 1-character level (hereinafter referred to as a single Kanji. The single Kanji referred to here is Hiragana. , Katakana, etc.) are included in the analysis. In addition, whether or not this single kanji can be a word (for example, whether it is registered as a one-character word in the Japanese dictionary or registered as a prefix or suffix) is registered in the word dictionary 11. Has been done. If one character cannot be a word, it is processed as an unknown word (details will be described later).

Next, the unknown word extracting means 23, the unknown word transmitting means 25, the information inputting means 27 and the information registering means 29 in the first embodiment will be described with reference to FIGS.
Here, FIG. 4 and FIG. 5 are flowcharts showing the operations of the respective means 23 to 29.

Upper device (text data generator) 10
When the text data is input from 0 (step S41 of FIG. 4), the text analysis unit 13 performs a morphological analysis process which is a part of the text analysis process on the input text data (step S42). For example, the input text data is referred to a word registered in the word dictionary 11 and morphologically analyzed by a known method such as the longest matching method. In this morphological analysis, a word that is neither a single Chinese character nor a prefix or suffix, that is, an unknown word is extracted (step 4 in FIG. 4).
3, in detail Fig. 5). In the first embodiment, all the extracted unknown words are stored in the unknown word buffer 23a (see FIG. 4).
Step S44). If it is not an unknown word, it is confirmed whether it is the sentence end (end of the input text data) (step S45 in FIG. 4). The sentence end here can be the end of a unit read aloud by voice. For example, it can be the end of a sentence, the end of a sentence, or the end of a phrase. If it is not the end of sentence, the morphological analysis and unknown word extraction processing for the following text data is continued (S42, S43). When the control section (unknown word extracting means 23) of the text-to-speech conversion system detects the sentence end, the control section (unknown word transmitting means 25)
Opens the unknown word buffer 23a (step S46 in FIG. 4) and collectively returns the unknown word morphemes in the unknown word buffer 23a to the upper level device 100 as they are. Host device 10
0 indicates the returned unknown word, for example, the input / output buffer 10
The data held in 1 is displayed on the screen of, for example, a CRT as one component of the unknown word transmission means (step S46 in FIG. 4), and a message prompting the user to input is displayed. At this time, that is, when the unknown word input is detected in the input / output buffer 101, if the higher-level device 100 notifies the user that the unknown word is displayed by a beep sound, a voice message, or the like, the user interface is more excellent. Needless to say, it becomes a system. Next, the user inputs the information necessary for registering the unknown word in the user dictionary 11b (see FIG. 3) of the word dictionary 11 from the information input means 27 (step S47 in FIG. 4). This information can be, for a morpheme detected as an unknown word, its reading, part of speech, accent position, connection information before and after, and the like. At this time, if an example of input is displayed on the screen, the system has a better user interface.

When the predetermined information is input by the user, the upper level device 100 sends the unknown words and the information (the reading, the part of speech, etc.) input by the user for the unknown words to the text-to-speech conversion device 200. In the text-to-speech conversion device 200, the information registration means 29 automatically registers these pieces of information in the user dictionary 11b (step S48 in FIG. 4). Here, the unknown word can be automatically registered in the user dictionary 11b by the following procedure, for example. This description will be described with reference to FIGS. Note that, here, the higher-level device 100 to the text-to-speech conversion device 2
00 (hereinafter, also referred to as a transmission buffer) for storing information to be sent to 00, and information (also referred to as a return value) returned from the text-to-speech conversion apparatus 200 to the upper level apparatus 100.
A buffer for storing (hereinafter also referred to as the return value buffer)
An example using and will be described. These buffers can be composed of, for example, the input / output buffer 201. The inside of the return value buffer contains the code of the morpheme that became the unknown word, and the EOF code at the end thereof (see FIG. 6). The number of unknown words contained varies depending on the number of extracted unknown words, and may be single or plural. When there are a plurality of unknown words, it is assumed that they can be distinguished by a line feed code "\ 0" indicating the division (also see FIG. 6).

FIG. 7 is a flow chart showing a process in which the higher-level device 100 reads an unknown word from the return value buffer and performs automatic registration. One character is read from the return value buffer shown in FIG. 6 (step S71) and it is determined whether the character is a line feed code (S72). The contents of the return value buffer are read until a line feed occurs, and the information before the line feed code is treated as an unknown word and information input processing such as part of speech (S73) is performed.
The data is stored in the transmission buffer (S74). FIG. 8 is a data structure diagram showing an embodiment of the transmission buffer. In the process of FIG. 7, when the host device detects the EOF code next to the line feed (S72, S75), the transmission buffer stores the information shown in FIG. Send to. These pieces of information are stored in the user dictionary 11b.

Next, phonological prosodic symbol live processing, voice synthesis parameter generation processing, and voice synthesis processing are performed, and voice output is performed (steps S49, S50, S51). The phonological prosodic symbol raw process, the voice synthesis parameter generation process, and the voice synthesis process can be performed by any suitable method. Since these processes are not the subject of the present invention, they will be briefly described. In phonological prosodic symbol string processing, based on the information stored in the word dictionary, independent words and independent words are connected, independent words and auxiliary words are connected to create bunsetsu, and bunsetsu and bunsetsu A phoneme that describes the input sentence reading, accent, intonation, pose, etc. as a character string while making the reading of the sentence by connecting and determining the strength of the pose position and accent position in the sentence at that time. Generate a prosodic symbol string (intermediate language). In the synthesis parameter generation process, the speech unit (sound type) data is retrieved from the speech unit storage unit 15 (see FIG. 1) based on the phonological prosodic symbol string,
The duration of each phoneme (sound length) is set, the pitch pattern (sound pitch) is determined based on the accent information of the phoneme prosodic symbol string, and the amplitude pattern (sound pitch) is determined by the length of each phrase. (Strength) is determined, and a voice synthesis parameter including these pieces of information is generated. Here, the speech unit is a basic unit for connecting and creating a synthetic waveform, and various units are prepared in advance according to the type of sound. The speech unit storage unit 15 stores such speech units. The voice synthesis unit 19 (see FIG. 1) generates voice waveform data based on the voice synthesis parameters. The audio output unit 21 (see FIG. 1) temporarily holds the audio waveform data in a built-in buffer memory, and performs digital / analog conversion of this data according to a predetermined clock to convert it into an analog audio signal. Then, the built-in speaker outputs this analog audio signal as a sound, or transmits it to another device by the transmitting means.

In the system of the first embodiment of the first aspect of the present invention, unknown words can be extracted for each input text data, and when the end of the input text data is reached, the extracted unknown words are automatically collected by the user. Further, the user can input information necessary for registering unknown words in the user dictionary, and the system can perform batch automatic registration of these unknown words in the word dictionary.

1-2. Second Embodiment of First Invention In the first embodiment described above, all unknown words extracted up to the end of the input text data are stored in the unknown word buffer and transmitted to the user even if they are duplicate unknown words. If this redundant storage and redundant transmission are eliminated, for example, the memory capacity and the processing time can be shortened, which is considered preferable. The second embodiment is an example of improving this. Therefore, in this second embodiment, the unknown word extracting means 23 is configured as follows. FIG. 9 is a diagram used for the explanation. In the second embodiment, the unknown word extraction means 23 determines whether an unknown word that is the same as the extracted unknown word is already stored in the unknown word buffer every time the unknown word is extracted from the input text data. Check (step S52 in FIG. 9). Then, if it is not stored, the extracted unknown word is stored in the unknown word buffer (step S53 in FIG. 9). If the same unknown word as the extracted unknown word is stored in the unknown word buffer, the unknown word is not stored in the unknown word buffer, and the process proceeds to the determination process of whether it is the sentence end or not. If the end of the sentence is detected, the processes from step S46 onward are performed as in the first embodiment.

1-3. Third Embodiment of First Invention In the above first and second embodiments, the user inputs the information (reading, part of speech, etc.) required when registering an unknown word in the word dictionary by the information input means. On the other hand, in the third embodiment, by providing means for accessing a large-scale database, the information necessary for registering an unknown word in the word dictionary may be obtained from the database in some cases. 10 and 11 are diagrams provided for the description.

In FIG. 10, 71 indicates a larger database in which information necessary for registering an unknown word in the word dictionary may be registered, and 73 is a network. The network 73 connects the text-to-speech conversion system 75 (three units are shown in FIG. 10) of the third embodiment and the large-scale database 71. The third embodiment shows an example in which the database access means 103 is provided on the higher-level device 100 side.

The standard dictionary 11a (see FIG. 3) existing inside the text-to-speech conversion apparatus 200 has only tens of thousands to hundreds of thousands of words registered therein, and therefore there is not much data such as proper nouns. On the other hand, the large-scale database (for example, large-scale Japanese database) 71 includes proper nouns,
There are data of hundreds of thousands to millions of words. The third embodiment can use such a database.
The procedure will be described below.

The unknown word extracted by the unknown word extracting means 23 on the side of the text-to-speech converter 200 is the unknown word transmitting means 2
5 is transmitted (input) to the higher-level device 100 (see FIG.
1, step S81), the host device 100 accesses the large-scale Japanese database 71 on the network 73 (step S82 in FIG. 11), and the text-to-speech conversion device 2
It is searched whether or not the information necessary for registering the word dictionary of the unknown word received from the 0 side exists in the database 71 (FIG. 1).
1 step S83). When the information of this unknown word exists in the database, the data is taken in (S84, S
85). When there is no information of this unknown word in the database, the user inputs information by the information input means 27 by the procedure described in the first and second embodiments (S86,
S87). Then, the user dictionary is automatically registered (S8
8), proceed with the analysis.

As a matter of course, the timing at which the host device 100 accesses the set work 73 may be for each unknown word or may be for all the unknown words stored in the unknown word buffer at once.

According to the system of the third embodiment, there is an advantage that the user may be able to reduce or eliminate the information input about the unknown word.

1-4. Fourth Embodiment of First Invention In the above-described first to third embodiments, the unknown word is described as a character. However, in the text data, J
Many so-called external characters (for example, a symbol for creating a chart or a special mark) to which a character character code according to a standard other than the standard character code defined by IS, shift JIS, EUC, etc. are given.
Such external characters become unknown words. For example, when text data created by a stand-alone word processor is input to the host device 100 of the text-to-speech conversion system via a medium, the external character unique to this stand-alone word processor becomes an unknown word. Further, since such an external character is expressed by a code different from the standard code in the system, it cannot be expressed by the host device of this text-to-speech conversion system. Therefore, in the system according to the fourth embodiment, when the unknown word is an external character and the standard code in the system of the external character exists, the code assigned to the external character is replaced with the standard code.
Equipped with external character code replacement means. This will be described in detail below with reference to FIG.

First, such an external character is extracted by the procedure of extracting an unknown word mainly described in the first embodiment, and the external character is displayed on the host device. In the dictionary of ordinary characters other than the external character, the dictionary corresponding to one morpheme is composed of notation, reading, part of speech, accent position, connection information, etc., as shown in FIG. This is created by inputting each information for the notation of the morpheme that has become an unknown word. However, when the unknown word is an external character, in this fourth embodiment, as shown in FIG.
As shown in (B), the external character is replaced with the symbol in the standard dictionary. That is, for the notation that became an unknown word, read,
The code in the standard dictionary is stored in the replacement code area without inputting the part of speech, the accent position, and the connection information. Then, it is registered in the user dictionary in the state of FIG.
Even the external characters are replaced with the standard dictionary code,
Morphological analysis can be performed on an external character even when an unknown word is input. Of course, the user does not need to know what is the code of the morpheme that has become an unknown word.

2. Description of Embodiment of Second Invention Next, an embodiment of the second invention will be described in order to facilitate the use of the user dictionary when a plurality of users use the same text-to-speech conversion system. The second aspect of the invention is characterized by including means for securing an area for each user in the word dictionary, password input means, dictionary selection means, and dictionary table correction means. These means are configured as described below in each of the following embodiments.

2-1. First Embodiment of the Second Invention In the first embodiment of the second invention, means for securing an area for each user in the word dictionary is realized by devising the word dictionary. FIG. 13 is a diagram showing the structure of the word dictionary 11. That is, the word dictionary 11 is the standard dictionary 1
1a, 11b ₁ , 11b ₂ , ..., 11b _N , which are N user dictionaries, and 11c ₁ , 11c ₂ ,.
It has _N dictionary tables indicated by 11c _N and a standard dictionary table 11d. Here, N corresponds to the number of users. The areas of the standard dictionary, the plurality of user dictionaries, and the plurality of dictionary tables are allocated in the memory of the system. Further, in this embodiment, a set including a user dictionary and a dictionary table for a user and a set of users other than this user are
The same address of the memory is redundantly assigned and selectively used. The allocation of these word dictionaries on the memory will be described later (will be described later with reference to FIG. 16).

Next, the password input means and the dictionary selection means will be described. FIG. 14 is a diagram used for the description. As described above, the host device 100 can be composed of, for example, a personal computer, a workstation, a word processor, or the like. In recent years, in order to maintain confidentiality with the expansion of information communication, it is common to use a code such as a password to confirm the system usage right before using the system from the host device 100 before using the system.
In this embodiment, the password input means 110 is installed in the host device.
Equipped with. Further, in this embodiment, the password input means 11
0 is a password input unit 110a and a password output unit 11
It is composed of 0b. The upper level device 100 uses the password as the password input unit 110a of the password input means 110.
And accept this password in the password output section 110
From b, dictionary selection means 2 in the text-to-speech conversion device 200
Send to 10. Dictionary selection means 21 in the case of this embodiment
0 indicates the password input unit 212 and the password identification unit 21.
4 and. In the dictionary selection means 210, the password input unit 212 receives the password and transfers the password to the password identification unit 214. Password identification unit 214
Then, the dictionary table corresponding to the password and the memory area of the user dictionary are secured in the word dictionary. Next, an embodiment for securing the memory area of the word dictionary unit according to the identification of the user code and the password will be described. FIG. 15 is a block diagram showing a specific configuration example of the password identification unit 214 described above. In this case, the password identification unit 214 is composed of a password register 214a for holding the password input here and a decoder unit 214b for generating a memory select signal according to the value held in the password register 214a. Password register 21
The bit length of 4a varies depending on the number of users (users), but is 16 bits when, for example, 256 people at maximum are used, that is, N = 256.
The output of the decoding unit 214b is generated by the address decoding of the CPU in the control unit when the control unit 201 (see FIG. 2) of the text-to-speech conversion device accesses the word dictionary 11. Then, the output of the decoding unit 214b becomes valid only when the chip select signal CS is active. As a result, the control unit has a plurality of dictionary tables 1
1c ₁ ~11c _N, without complicated control-conscious user dictionary 11b ₁ ~11b _N, can be accessed dictionary table and user dictionary a simple process similar to dealing with only the dictionary unit.

FIG. 16A is a memory map showing the structure of the memory in the dictionary section, which is controlled by the chip select (memory select) signal produced by the decoding section 214a in the dictionary selecting means 210. The example of FIG. 16A shows an example in which user dictionaries for 256 people and a dictionary table are stored. The physical addresses of the user dictionary and dictionary table shown in FIG. 16 will be described in detail below. FIG. 16B shows an example thereof (memory map of the CPU in the control unit 201 (see FIG. 2)). In this example, 0000
Program area of CPU from address h to 3FFFh, CPU from address 4000h to 7FFFh
Work area, from 8000h to BFFF, the dictionary table area and the user dictionary area (here, 8
A dictionary table area is provided from addresses 000h to 9FFFh, and a user dictionary area is provided from addresses A000h to BFFFh. ), FFF from C000h
Up to the address Fh, the standard dictionary area and the standard dictionary table area are provided. Therefore, the CP in the control unit
It can be seen that the address decoding of U can divide the memory into the above-mentioned four areas by decoding the upper 2 bits of the address. For example, the upper 2 bits are "1,
When it is "0,", a peripheral circuit of the CPU outputs a chip select signal for accessing the user dictionary and the dictionary table. Since the 16-bit configuration is used here, the upper 2 bits are 1. Since it is 0, the highest hexadecimal display is any one of 8, 9, A, and B, which means that any of the addresses from 8000h to BFFF can be accessed. The size of each address is 14 bits (3 fffh), which is obtained by subtracting 2 bits used for address decoding from the total number of bits of the address. The upper 2 bits of the address of the control unit CPU described above have a plurality of areas of "1, 0," (having the same address in duplicate. Corresponds to things (but are used by switching as described above), each user dictionary, the size of the dictionary table, as described above, the size of 14bit length.

FIG. 17 is a diagram collectively showing an example of the hardware configuration of the dictionary selecting means. That is, it is a hardware example of a dictionary selection means including a plurality of user dictionaries, a plurality of dictionary tables, an address decoding unit around the control unit CPU of the text-to-speech conversion apparatus 200, a password register, and a password decoder. According to this configuration, in the system having a plurality of user dictionaries in the text-to-speech conversion device, even if the user changes,
By simply inputting a password, it is possible to realize an apparatus for performing morphological analysis by using a user dictionary corresponding to a password of a user who is automatically registered.

By the way, the dictionary table 11 in the word dictionary
c is for efficiently performing morphological analysis with reference to a dictionary group including a standard dictionary and a user dictionary,
For example, the configuration is as shown in FIG. That is, one character of the text data input to the text-to-speech conversion device is registered as the key code 91 in the dictionary table. Further, in the dictionary table, the address 93 of the standard dictionary or the user dictionary in which the longest morpheme (word) starting with the key code exists is registered. By tracing this address, the morpheme of the first candidate starting with the first character of the input text data is retrieved from the dictionary and is matched with the input text data. Needless to say, if none of the user dictionaries in the word dictionary are registered, all dictionary addresses in the dictionary table refer to the standard dictionary area.

In such a dictionary table, (A): when a new user dictionary is downloaded from the external medium (floppy disk) to the user dictionary area, and (B): a new word is added to the user dictionary when the system is used thereafter. When registering (what was an unknown word), it is necessary to modify the dictionary table. This is because the dictionary table cannot be created unless the key code and its address are changed in response to the contents of the user dictionary changed. There are various possible methods for modifying the dictionary table. An example will be described below. 19 and 20 are diagrams used for the description. In the following description, in order to distinguish between the dictionary table and the standard dictionary table, a dictionary table that is not the standard dictionary table may be referred to as a “user dictionary table”.

FIG. 19 is a flow chart showing an example of a procedure for correcting the user dictionary table, and FIG. 20 is an explanatory diagram of links between pieces of information of each word in the word dictionary.

The text-to-speech conversion device 200 (dictionary table correction means) secures a memory area for the user dictionary corresponding to the password sent from the upper level device 100,
A user dictionary from, for example, an external medium is downloaded there (step S1 in FIG. 19). Further, a process (link process) of scanning the notations in which the code of the first letter of the notation in the user dictionary is the same from the longest notation is performed (step S2). Further, at this time, it is determined whether or not there are some links (the number of links L) in which the code of the first character is the same in the user dictionary and this is stored in the buffer (step S).
3). Here, the link process and the number of links will be described with reference to the example of FIG. 20. When linking data whose first character is “Oki”, “Oki Electric” at the physical address A000h of the user dictionary is scanned first and then In addition, “Oki □ in FIG.
The address is linked to the address at the right end of the "Electrical" column (this is the address information registered to indicate the link destination), that is, the physical address A002h is scanned. In addition, data in which the first character of the notation is "Gon" is linked. In this example, there are links for "Oki" and "On", so the number of links L is 2. Link processing is done in this way, and
When the number of links L is obtained, the text-to-speech conversion device 200
The (dictionary table correction means) sets the variable n to 0 (step S4) and copies the data of the standard dictionary table to the user dictionary table (step S5). Next, the variable n is incremented (step S6), and then, from this standard table, the character of the number of links indicated by the variable n, specifically, if the number of links is 2, first The address of the data that matches the character notation of the link (in the above example, the character notation of the link "Oki") is taken out (step S7). Next, the address of "Oki" in the standard table is the address field at the right end of the last data field (the last word of the link, that is, the shortest morpheme data field) in the link of the first character "Oki" in the user dictionary Address column) (step S8). The process of step S8 will be described in detail with reference to the example of FIG. 20. In this case, the last data in the link in which the first character of the user dictionary is “Oki” is “Okinawa” at the physical address A002h. The address C000h of "Oki" in the standard dictionary is written in the rightmost address of the column. Next, the physical address (A000h in the case of FIG. 20) in the link in which the first character is "Oki" is written in the address field of the key code "Oki" in the user dictionary table (step S9). By the processing up to this point, the address change of the dictionary table for the key code "Oki" is completed. However, here, the address shown in FIG. 18 remains (A00
(It remains 0). Next, it is determined whether the variable n matches the number of links L (step S10). In this determination, steps S7 to S9 are performed for each link in the user dictionary (in the example of FIG. 20, the link of "Oki" and the link of "gratitude").
It is determined whether or not the process of (1) has been performed. That is, when n does not match L, the dictionary table is corrected for the remaining links (links in which the first character is "Gaku" here). The dictionary table can be corrected or changed by performing this processing to the end of the registered user dictionary.

2-2. Second Embodiment of Second Invention The system of the first embodiment of the second invention described above has a form in which a plurality of sets each including a user dictionary and a dictionary table are included in the word dictionary. On the other hand, in the second embodiment, even if there are a plurality of user dictionaries, a single dictionary table is provided, and the user dictionary corresponding to the user is automatically selected from the plurality of user dictionaries by the password entered by the user. 2 is an example of a system that is selectively selected.

FIG. 21 is a diagram showing the structure of the word dictionary in this second embodiment. This word dictionary 11 is a standard dictionary 1
1a, N user dictionaries 11b _{1 to} 11b _N , a single dictionary table (sometimes called a user common dictionary table) 11c, and a standard dictionary table 11d. Further, FIG. 22 shows a memory configuration for the user dictionary when the number of users N is 256, and the control unit 2
The memory map as seen from the CPU in 01 (see FIG. 2) is shown in FIG. As shown in FIG. 23, in the second embodiment, the program area of the CPU from the address 0000h to the address 3FFFh, the address from the address 4000h to the address 7F.
CPU work area up to address FFh, dictionary table area A0 from address 8000h to 9FFFh
Addresses 00h to BFFFh are user dictionary areas, and addresses C000h to FFFFh are standard dictionary areas. Therefore, it is understood that the address decoding of the CPU in the control unit can divide the memory into the above-mentioned five areas by decoding the upper 3 bits of the address. For example, when the upper 3 bits are "1, 0, 0," a chip select for accessing the dictionary table area is output by the peripheral circuit of the CPU, and "1, 0, 0" is output.
When 1, ", the chip select for accessing the user dictionary area is output. Here, 16 is selected.
Since it has a bit configuration, the upper 3 bits are 1, 0, 0
Therefore, since the highest hexadecimal display is either 8 or 9, it is possible to access any of addresses 8000h to 9FFF, and the upper 3 bits are 1.
Because it is 0 or 1, the highest hexadecimal display is A or B.
This is because any of the addresses 0h to BFFFh can be accessed. The physical size of each area of the dictionary table and the user dictionary is 13 bits (1 fffh) obtained by subtracting 3 bits for address decoding from the total number of bits of the address. Having multiple user dictionaries means C
The upper 3 bits of the PU address correspond to having a plurality of areas of "1, 0, 1,", and the size of each user dictionary is 13 bits long.

In the second embodiment, the user is identified by inputting a password, the memory area is secured, details of dictionary selection (however, in the second embodiment, selection of user dictionary), text-to-speech conversion device, control unit The dictionary decoding method viewed from the CPU is the same as that of the first embodiment of the second invention, and therefore the description thereof is omitted.

In the case of the second embodiment of the second invention, it is necessary to correct the dictionary table every time the system user changes. Since this correction can be performed by the same method as in the case of the first embodiment of the second invention, this explanation is also omitted.

2-3. Third Embodiment of Second Invention Next, in the case where a plurality of users use the same text-to-speech conversion system, and by devising that a user identifier is used, a single user dictionary can be shared by a plurality of users. An example will be described. Specifically, when a plurality of users use the same text-to-speech conversion system, a word that the user wants to register in the user dictionary has a user identifier that can be searched later by the password of the user (linked to the password). This is an example in which a plurality of users can perform morphological analysis using the words of the user dictionary that are individually registered by adding and creating a dictionary table individually with this user identifier.

FIG. 24 is a diagram showing the structure of the word dictionary in this third embodiment. This word dictionary 11 is a standard dictionary 1
And 1a, (also referred to as a composite user dictionary.) Single user dictionary 11b and, N pieces of user dictionaries Te ~ table 11c ₁ ~
11c _N and a standard dictionary table. Also,
FIG. 25 shows the memory configuration for the user dictionary table when the number of users N is 256, and the control unit 20
2 shows a memory map viewed from the CPU in FIG. 1 (see FIG. 2).
6 respectively. The memory map viewed from the CPU is basically the same as that of the second embodiment. However, in the second embodiment, a plurality of user dictionary areas are redundantly assigned to the same address, whereas in the third embodiment, a single user dictionary is used and a plurality of user dictionary table areas have the same address. The second and third embodiments are different in that they are assigned redundantly.

Next, the user identifier will be described. In the present embodiment, since the user dictionary itself is single, it is necessary to know which user registered each word in the user dictionary.

FIG. 27A shows the internal format of the user dictionary according to the third embodiment. The notation, reading, part-of-speech, accent position, connection information, and next candidate address information have already been described and will be omitted. However, the next candidate address is the address shown at the right end in FIG. 20, and is the address showing the link destination. Figure 2
The information of the user identifier in 7 (A) is the information newly set in the third embodiment.

As described above, it is premised that the user of the system inputs the password before using the system and then uses the password. In other words, while a user is using this system, the password is
0 and the text-to-speech converter 200 (see FIG. 2) are easy to know. When creating a user dictionary on this system, it is easy to write the user identifier linked with this password in the user identifier area of the user dictionary data. This writing of the user identifier is performed by the host device 1
It is clear that both sides can write the password, either 00 or the text-to-speech converter 200, because both can know the password.

FIG. 27B is a diagram showing an internal structure of an example of a user dictionary created using a user identifier. One column indicates one word, and the user identifier shown in FIG. 27A is stored at the end of one word. Here, "0
This is an example in which the code indicated by 0 "," 01 "," 02 "," 03 ", etc. is used as the user identifier indicating the user. The end of the user dictionary contains the EOF code indicating the end of the dictionary. There is.

With the user dictionary having the form as shown in FIG. 27, when a plurality of users add to the same dictionary, it is sufficient to add the registered word to the end of the dictionary. Unlike the case of managing individual dictionaries, unified management can be performed. Even if a user registers or deletes the word of his or her user identifier, the physical addresses of other dictionaries do not change, so only the dictionary table of the user who has made the change needs to be updated.

In the case of the third embodiment, it is necessary to correct the user dictionary table. The correction work of the user dictionary table can be basically performed by the method described in the first embodiment. However, step S2 shown in FIG.
The link processing work in is performed by the compound user dictionary 11b (see FIG. 24).
Refer only to the data selected by the user identifier in.

As is apparent from the above description, according to the third embodiment, even if another user deletes or adds a user dictionary, it is not necessary to change his table.
It can be seen that it is possible to construct a system that can be handled even when a plurality of upper-level devices simultaneously access the user dictionary of one text-to-speech conversion system.

2-4. Fourth Embodiment of Second Invention In the above-described first to third embodiments, at least one of the user dictionary and the dictionary table is N in number according to the number N of users who use the system. However, if the third embodiment is further developed, it is possible to use a single user dictionary and a single dictionary table. This fourth embodiment is such an example. The details will be described below.

FIG. 28 is a diagram showing the structure of the word dictionary in the fourth embodiment. This word dictionary 11 is a standard dictionary 1
1a, a single user dictionary (however, compound user dictionary described in the third embodiment) 11b, and a single dictionary table (however, user common dictionary table described in the second embodiment) 11
c and a standard dictionary table. The identification of the user by the password, the securing of the memory area, the process of assigning the user identifier in the registration of the user word, and the like can be performed by the corresponding methods of the first to third embodiments.

Next, the dictionary table and the user dictionary viewed from the CPU in the control unit 201 in the fourth embodiment will be described. As described above, in the present embodiment, there is only one compound user dictionary registered by a plurality of users and one user common dictionary table that is modified and updated according to changes in users. Memory allocation by the address decoder of the CPU seen in the embodiment does not occur, and only one memory area is required as a set of the user common dictionary table and the composite user dictionary.

Regarding the composite user dictionary, the composite user dictionary used in this embodiment is the same as the composite user dictionary described in the third embodiment (see FIGS. 27A and 27B). Therefore, its explanation is omitted. Further, the user common dictionary table needs to be modified every time the user changes, and even when the same user adds a user word during the text-to-speech conversion process. Such modification of the user S common dictionary table is performed by the modification method of the first embodiment, which considers only the user word selected by the user identifier (hence, the modification method of the third embodiment). You can do it.

From the above description, according to the fourth embodiment,
Since a single compound user dictionary and a single user dictionary table are sufficient, the user dictionary area can be variable in length, and a system with good memory usage efficiency can be constructed.

2-5. Fifth Embodiment of the Second Invention This fifth embodiment uses the user identifier more effectively. That is, when a single user dictionary is used by a plurality of users, and when user words are registered in the user dictionary by giving user identifiers to the user words, the same words are registered by different users. This is an example of using an identifier (shared user identifier) that can be shared by these different users if the data to be given to the morpheme such as notation and reading is the same.
This use means, for example, when a certain user-registered word also exists as a registered word of another user in the user dictionary area (of course, the identifier of the other user is given), the user identifier area of that word is displayed. It is used to rewrite the stored other user's identifier to the shared user identifier. By doing so, it is possible to prevent the same user word having the same morpheme data from being registered in duplicate, so that it is possible to further improve the memory use efficiency. The details will be described below.

FIG. 29A is a diagram showing a code example of a user identifier (user identifier in the third and fourth embodiments) corresponding to the reference example in the fifth embodiment. Specifically, an example is shown in which codes of unique user identifiers of 16 users in total of users 1 to 16 are formed with a 4-bit structure. On the other hand, FIG. 29B is a code example of the shared user identifier in the fifth embodiment. Specifically, four users can be identified by 4 bits, and the user word can be registered in a state where it can be shared by two to four people, as well as when the user word is registered by each user. FIG. 29
In (B), the code “0001” is an example in which only the user D has registered the user word, the code “1100” is shared by the users A and B, and the code “1101” is the users A, B and An example shared by D, the code of "1111" is A
It is an example of sharing with four people of ~ D. According to the code shown in FIG. 29B, it can be seen that the registered word can be shared by operating each bit. When creating or changing the dictionary table based on this, pay attention only to the bits of the user identifier data of the user corresponding to the password input from the upper level,
If this bit is 1 if it is valid, and if it is 0 it is invalid, then the dictionary table may be created in the same manner as in the third and fourth embodiments. Next, regarding the registration of the user word in the user dictionary in the fifth embodiment, before writing the word registered by the CPU in the control unit in the user dictionary area, the dictionary data of the word to be registered and the already registered word are already registered. The already-combined user dictionary is subjected to a compare check, and if there is the same, the user identifier bit of the registered word is set to 1, and the user identifier bit of the user of the word to be registered is set to 1. If it has not been registered, the user identification bit of the user to be registered in the user identifier area of the word to be registered may be registered as 1.

3. Description of Embodiment of Third Invention Next, an embodiment of a text-to-speech conversion system having a function of notifying the user of the fact that the user dictionary in the word dictionary is changed or added while the system is being used will be described. . The text-to-speech conversion system according to the third aspect of the invention includes a user dictionary change detection unit for detecting whether or not the information in the user dictionary in the word dictionary is changed during the text-speech conversion work, and the user dictionary is changed. In this case, the user dictionary change transmitting means for notifying the user of this is provided. Each of these means is configured as described below in each of the following embodiments.

3-1. First Embodiment of Third Invention FIG. 30 is a block diagram showing a main part of a text-to-speech conversion system according to a first embodiment of the third invention. In the first embodiment, the user dictionary change detecting means 120 compares the sizes of the data in the user dictionary at the start of use of the system and at the end of use of the system, and displays a bit indicating the total amount of this data. This is an example of a configuration in which the least significant 1 bytes are compared with each other. Specifically, it has the following configuration. The data of the user dictionary will be described as being stored in the user dictionary storage medium 121 owned by the user (for example, a floppy disk or a memory in the host device).
Although not limited to this, here, an example in which the user dictionary change detection means 120 is provided on the higher-level device 100 side is shown.

When the host device 100 writes the data of the user dictionary to be registered from the user dictionary storage medium 121 to the user dictionary 11b of the word dictionary 11 of the text-to-speech conversion device 200, the size of the user dictionary and the contents of the user dictionary. The lower 1 byte of the result of adding all the data of
Called BCC data. ) And the first storage means 123
(Hereinafter, referred to as the first size buffer 123). The first size buffer 123 stores only the size of the user dictionary at the time of initial registration, and is not rewritten when the user dictionary already registered is added or changed. Next, when the system ends, the text-to-speech converter 20
The control unit 201 of 0 measures the size and the BCC data of the user dictionary in the word dictionary 11, and the size and the data are transmitted from the input / output unit via the input / output unit of the host device.
Send to 00. The host device 100 has these sizes and BCC
The data is written in the second storage means 125 (hereinafter referred to as the second size buffer 125). Note that the size of the user dictionary is indicated by the EOF code at the end of the user dictionary as described above, so the size up to the EOF code is the size of the user dictionary. Also, since the first size buffer 123, the second size buffer 125, and the user dictionary size buffer in the word dictionary are all cleared to default zero (initialized with "0" data), it is necessary to register the user dictionary. If there is no such data, the user dictionary, the first size buffer, and the second size buffer are all kept as "0" data.

Upon notifying the text-to-speech conversion device of the system termination processing, the upper device receives the first size buffer 123.
Is not "0", waits for the data to be written in the second size buffer 125, and when written, the size data of the first size buffer 123 and the size data of the second size buffer 125 and the BCC data are respectively written. If either of them is compared, it is determined that the user dictionary has been changed or modified between the start and end of system use, and a message is displayed on the CRT of the host device or the dictionary is changed to the user by voice output. Notify that it has been fixed.

Next, an example of the user dictionary change transmitting means for notifying that the user dictionary has been changed or modified will be described. When preparing a character string such as the message example "Your user dictionary has been updated." In the upper device and outputting it to the screen stored in the memory of the upper device, output the character string in the program. Is displayed on the screen, and when outputting the voice, the text-to-speech conversion device may be instructed to be the output destination. Of course, it is possible to output to both.

3-2. Second Embodiment of Third Invention In the first embodiment of the third invention described above, the detection of the dictionary update is
Although the comparison is performed by comparing the dictionary size and the BCC data, other configurations may be used. Therefore, in the second embodiment, as shown in FIG. 31, the dictionary status register 127 in which the bit data is changed when the user dictionary is changed is used, and whether the user dictionary is changed or not based on the bit data change of the register 127. Is detected. Below, FIG.
The second embodiment will be described with reference to FIGS. Here, FIG. 31 is a block diagram showing the main part of the text-to-speech conversion system of the second embodiment, and FIG. 32 is a diagram showing an example of the configuration of the dictionary status register.

When writing the data of the user dictionary to be registered from the user dictionary storage medium 121 into the user dictionary area of the word dictionary 11 of the text-to-speech conversion device 200, the upper level device 100 sets the user dictionary presence / absence bit of the dictionary status register. Set "1" to (see FIG. 32). It is assumed that the dictionary status bits are cleared to zero when the system starts to be used. Further, when there is no user dictionary to be registered, the user dictionary presence / absence bit of the dictionary status register is not set as "0". Further, when the system is started with the user dictionary already registered, the user dictionary presence / absence bit of the dictionary status register remains "0" although it is different from the actual situation.
Next, when the user dictionary is modified or added while the system is being used, the host device corrects or corrects the dictionary status register when the data of the additional user word is transmitted to the text-to-speech conversion device, and the added bit has "1". "Set up." If the user dictionary existence / non-existence bit is "0" when "1" is added to the correction / addition bit, this bit is also set to "1" at this time. In other words, the values that the dictionary status register can take are "0,0,""1,
There are three types: 0, "" 1,1, "(" User dictionary presence / absence bit, correction addition bit, 1: Yes, 0: No) ".
The pattern of 1, "cannot exist. Of course, the bit configuration of the dictionary status register is not limited to this and can be a suitable one.

Next, at the end of the system, the host device refers to the correction addition bit of the dictionary status register, and when this value is "1", determines that the user dictionary has been updated,
For example, a message is output to the upper CRT or a voice is output to notify the user that the dictionary has been changed or corrected. Details of the transmission method have been described in detail in the first embodiment, and will not be repeated.

3-3. Third Embodiment of Third Invention In the first and second embodiments of the third invention described above, the fact that the user dictionary has been updated can be communicated to the user. Therefore, the contents of the user dictionary currently used by the user It was possible to prompt whether or not it was necessary to back up to the user dictionary storage medium. However, it is more preferable in terms of protection of the user dictionary if the contents of the user dictionary currently used can be automatically backed up in the user dictionary storage medium. This third
The example is an example. This description will be given with reference to FIGS.

Detecting that the user dictionary has been updated and transmitting this point to the user can be performed as in the first or second embodiment. In the third embodiment, the user is informed that the user dictionary has been updated, and the text-to-speech conversion system 2 is sent from the host device 100.
00 to send a user dictionary backup command. The control unit 201 of the text-to-speech conversion system 200 receives the user dictionary backup command. On the other hand, the control unit 210 accesses the user dictionary area of the word dictionary 11 and indicates E from the beginning of the user dictionary to the end of the user dictionary.
Text-to-speech conversion device 20 that reads up to OF code
It is output to the higher-level device 100 from the 0 input / output unit. The upper level device 100 overwrites the user dictionary data returned from the text-to-speech conversion device 200 on the user dictionary storage medium 121, and when this overwriting is completed, the backup of the user dictionary is completed and the system termination is completed.

As described above in detail, when the system is terminated, the user dictionary is corrected and additions are confirmed. When the corrections or additions are made, a message is output to the user and a text voice for automatically backing up the user dictionary. A conversion system can be realized.

3-4. Fourth Embodiment of Third Invention The text-to-speech conversion system of the fourth embodiment is the system of the third embodiment with the addition of a function for selecting whether or not to perform automatic backup.
This will be described with reference to FIGS.

Detecting that the user dictionary has been updated and transmitting this point to the user can be performed as in the first or second embodiment. In the fourth embodiment, the user is informed that the user dictionary has been updated, and the user is informed of whether or not to automatically backup the message, for example, "Your user dictionary has been updated. If yes, enter "y". If not, enter "n". "CRT screen display or voice output is performed. At this time, the high-level equipment 100 waits for the code "y" or "n" input by the user, and waits for "y".
In the case of, the system is terminated by performing the automatic backup shown in the third embodiment, and when the input from the user is "n", the system is terminated without performing the automatic backup of the user dictionary. As can be seen from the above description, the present invention can realize a text-to-speech conversion system having a better user interface.

4. Description of Fourth Invention Although not described in detail, a text-to-speech conversion apparatus that combines the first to third inventions is used to obtain a system having the effects of the first to third inventions.

[0081]

As is apparent from the above description, according to the text-to-speech conversion system of the first invention of this application, unknown words included in the input text data up to the end are sequentially extracted at the stage of morphological analysis processing. Means, and means for collectively transmitting the unknown words extracted so far to the user when the end of the input text data is reached, and the information necessary for registering the transmitted unknown words in the word dictionary. Means for inputting an unknown word and means for registering an unknown word and information input about the unknown word in a word dictionary. It can be transmitted later, and the user can also input information in a batch. Therefore, the above problems (1) to (4) can be solved. In addition, in the case where the unknown words are sequentially extracted, sequentially transmitted, and the information is sequentially input, it is considered that, for example, the user may be restricted by the system.
The present invention alleviates that point. For example, in the present invention, it is possible to perform unknown post-extraction at night or during breaks. Further, since the unknown word is extracted at the stage when the morphological analysis result is obtained, the unknown word extraction processing time can be shortened as compared with the case where the unknown word is extracted after the process of adding the temperament and prosody.

Further, according to the text-to-speech conversion system of the second invention of this application, a means for securing an area for each of a plurality of users in at least the user dictionary of the word dictionary and a password for identifying the plurality of users are provided. Since the means for inputting, the dictionary selecting means for selectively enabling the dictionary area corresponding to the password input from the password inputting means, and the means for correcting the dictionary table are provided,
A user dictionary can be dedicated to each user. This eliminates the need for the old user to back up the user dictionary from the system as the user changes, while the new user downloads his or her user dictionary to the system.

According to the text-to-speech conversion system of the third invention of this application, the means for detecting whether or not the information in the user dictionary is changed during the text-to-speech conversion work, and the user dictionary are changed. Since it is provided with a means for notifying the user of that fact, it is possible to prompt the user to back up the user dictionary. Therefore, the risk of the user forgetting to back up the user dictionary can be reduced.

[Brief description of drawings]

FIG. 1 is a diagram mainly used for explaining a system of a first invention.

FIG. 2 is a view following FIG. 1 mainly for explaining the system of the first invention.

FIG. 3 is a diagram showing a configuration example of a word dictionary in the first invention.

FIG. 4 is a procedure explanatory diagram of the first embodiment of the first invention.

FIG. 5 is a procedure explanatory view (No. 2) of the first embodiment of the first invention.
Is.

FIG. 6 is an explanatory diagram of information registration means.

FIG. 7 is an explanatory view of the information registration means following FIG. 6;

8 is an explanatory diagram of the information registration means following FIG. 7. FIG.

FIG. 9 is an explanatory diagram of a second embodiment of the first invention.

FIG. 10 is an explanatory diagram of the third embodiment of the first invention.

FIG. 11 is an explanatory diagram of the third embodiment of the first invention and is an explanatory diagram of data access means.

FIG. 12 is an explanatory diagram of external character code replacing means.

FIG. 13 is an explanatory diagram of the first embodiment of the second invention.

FIG. 14 is an explanatory diagram of the first embodiment of the second invention, following FIG.

FIG. 15 is an explanatory view of the first embodiment of the second invention, following FIG.

FIG. 16 is an explanatory view of the first embodiment of the second invention, following FIG. 15;

FIG. 17 is an explanatory view of the first embodiment of the second invention, following FIG. 16;

FIG. 18 is an explanatory diagram of a dictionary table.

FIG. 19 is an explanatory diagram of a dictionary table correction means.

FIG. 20 is an explanatory diagram of data links in the word dictionary.

FIG. 21 is an explanatory diagram of the second embodiment of the second invention.

FIG. 22 is an explanatory diagram of the second embodiment of the second invention, following FIG. 21;

FIG. 23 is an explanatory diagram of the second embodiment of the second invention, following FIG. 22;

FIG. 24 is an explanatory diagram of the third embodiment of the second invention.

FIG. 25 is an explanatory view of the third embodiment of the second invention, following FIG. 24;

FIG. 26 is an explanatory diagram of the third embodiment of the second invention, following FIG. 25;

FIG. 27 is an explanatory diagram of the third embodiment of the second invention.

FIG. 28 is an explanatory diagram of the fourth embodiment of the second invention.

29A is an explanatory diagram of a comparative example in the fifth embodiment, and FIG. 29B is an explanatory diagram of a code example of a shared user identifier which is an explanatory diagram of the fifth embodiment.

FIG. 30 is an illustration mainly of the first embodiment of the third invention.

FIG. 31 is an illustration mainly of the second embodiment of the third invention.

32 is an explanatory diagram of the third aspect of the invention, mainly following the FIG. 31. FIG.

[Explanation of symbols]

11: word dictionary 11a: Standard Dictionary 11b: user dictionary (which may be a composite user dictionary) 11b ₁ ~11b _N: user dictionary 11c: dictionary table (user dictionary table, there is a case of the user common dictionary table) 11c ₁ ~ 11c _N : dictionary table (user dictionary table) 11d: standard dictionary table 13: text analysis unit 15: speech element storage unit 17: synthesis parameter generation unit 19: speech synthesis unit 21: speech output unit 100: upper device 200: text Voice converter

Claims (16)

[Claims] 1. A text analysis unit for generating a phoneme and a prosodic symbol string from text data by a process including a morphological analysis process performed by referring to a word dictionary and a subsequent phoneme / prosodic process, and the generated phoneme. , A synthesis parameter generation unit for converting a prosodic symbol string into a voice synthesis parameter using a voice unit storage unit, a voice synthesis unit for converting the generated voice synthesis parameter into voice waveform data, and the generated In a text-to-speech conversion system that has a speech output unit that converts speech waveform data into analog speech signals and outputs the unknown speech, unknown words that are included in the input text data up to the end are sequentially extracted at the stage of morphological analysis processing. An extracting unit, an unknown word transmitting unit that collectively conveys to the user the unknown words extracted so far when the end of the input text data is reached, Information input means for the user to input the necessary information for registering the obtained unknown word in the word dictionary, and information registration for registering the unknown word and the information input for the unknown word in the word dictionary A text-to-speech conversion system comprising means. 2. The text-to-speech conversion system according to claim 1, wherein the unknown word extracting means includes an unknown word buffer for storing the unknown word, and the unknown word is extracted from the input text data. Is configured to check whether the same unknown word as the extracted unknown word is already stored in the unknown word buffer, and if not, store the extracted unknown word in the unknown word buffer. The text-to-speech conversion system is characterized in that the unknown word transmitting means is configured to transmit an unknown word stored in an unknown word buffer to a user. 3. The text-to-speech conversion system according to claim 1, wherein a necessary information for registering an unknown word in a word dictionary is registered, and a larger database is accessed to access the necessary information. A text-to-speech conversion system having a database access means for importing. 4. The text-to-speech conversion system according to claim 1, wherein when the transmitted unknown word is an external character and a standard code of the external character in the system exists, a code assigned to the external character. A text-to-speech conversion system characterized by comprising external character code replacement means for replacing a standard character with a standard character. 5. A text analysis means for generating a phoneme and a prosodic symbol string from text data by a process including a morphological analysis process performed by referring to a word dictionary and a subsequent phonological / prosodic process. A synthesis parameter generating means for converting a phoneme and a prosodic symbol string into a voice synthesis parameter using a voice segment storage unit, a voice synthesis unit for converting the generated voice synthesis parameter into voice waveform data, and the generated voice synthesis data. A text-to-speech conversion system comprising a voice output unit for converting the voice waveform data into an analog voice signal and outputting the analog voice signal, wherein the word dictionary is a standard dictionary in which predetermined words are registered, and a system according to the use of the system. Text having a configuration having a user dictionary updated and modified by the user and a dictionary table for searching words from these standard dictionary and user dictionary In the speech conversion system, means for securing an area for each of a plurality of users in at least the user dictionary of the word dictionary, password input means for inputting a password for identifying the plurality of users, and A dictionary selecting unit for selectively validating an area corresponding to the password input from the password input unit out of the areas for each of the plurality of users; and a unit for correcting the dictionary table. A text-to-speech conversion system characterized by. 6. The text-to-speech conversion system according to claim 5, wherein the means for ensuring an area for each user in the word dictionary includes a standard dictionary, an individual user dictionary for each of a plurality of users, and a plurality of users. And an individual dictionary table according to the input password, among the plurality of individual user dictionaries and individual dictionary tables, the dictionary selecting means is effective. And the dictionary table correction means is configured to correct the individual dictionary table corresponding to the valid individual user dictionary when the information of the validated individual user dictionary is changed. Voice conversion system. 7. The text-to-speech conversion system according to claim 5, wherein the means for securing an area for each user in the word dictionary includes a standard dictionary, an individual user dictionary for a plurality of users, and a single user dictionary. And a dictionary table, and the dictionary selection means is configured to enable an individual user dictionary corresponding to the input password among the plurality of individual user dictionaries, and the dictionary table correction means. Is configured to correct the single dictionary table so as to correspond to the validated individual user dictionary. 8. The text-to-speech conversion system according to claim 5, wherein the means for securing an area for each user in the word dictionary is a word that the user wants to register in the user dictionary, and the password of the user is used. A text-to-speech conversion system, characterized in that the text-to-speech conversion system is a user identifier assigning unit for assigning a user identifier that can be searched later. 9. The text-to-speech conversion system according to claim 5, wherein the word dictionary has a standard dictionary, a single user dictionary, and an individual dictionary table prepared for each of a plurality of users, The means for securing an area for each user in the word dictionary is the user identifier assigning means according to claim 10, and the dictionary selecting means is the input password of the plurality of individual dictionary tables. The text-to-speech conversion system is characterized in that the individual dictionary table according to the above is enabled, and the dictionary table correction means is configured to correct each individual dictionary table when the information of the user dictionary is changed. 10. The text-to-speech conversion system according to claim 5, wherein the word dictionary has a standard dictionary, a single user dictionary, and a single dictionary table, and the user is included in the word dictionary. The means for ensuring each area is the user identifier giving means according to claim 10, and the dictionary table correction means is configured to correct the dictionary table when the user changes the information of the user dictionary. Characteristic text-to-speech conversion system. 11. A text analysis means for generating a phoneme and a prosodic symbol string from text data by a process including a morpheme analysis process performed by referring to a word dictionary and a subsequent phonological / prosodic process. A synthesis parameter generating means for converting a phoneme and a prosodic symbol string into a voice synthesis parameter using a voice segment storage unit, a voice synthesis unit for converting the generated voice synthesis parameter into voice waveform data, and the generated voice synthesis data. In a text-to-speech conversion system equipped with a speech output unit that converts speech waveform data into an analog speech signal and outputs it, it is detected whether the information in the user dictionary in the word dictionary has been changed during the text-speech conversion work. And a user dictionary change transmission means for notifying the user of a change in the user dictionary. Text-to-speech conversion system which is characterized. 12. The text-to-speech conversion system according to claim 11, wherein the user dictionary change detection means is the size of data in the user dictionary at the start and end of use of the system, and the total amount of each data. And a text-to-speech conversion system configured to detect whether or not the user dictionary has been changed by comparing at least two pieces of partial data of the total amount with the same parameters. 13. The text-to-speech conversion system according to claim 11, wherein the user dictionary change detection means has a dictionary status register in which bit data is changed when the user dictionary is changed. A text-to-speech conversion system having a configuration for detecting whether or not a user dictionary has been changed. 14. The text-to-speech conversion system according to claim 11, further comprising backup means for backing up the information in the user dictionary to a predetermined medium when a change in the user dictionary is detected. system. 15. The text-to-speech conversion system according to claim 14, further comprising backup execution / stop means for allowing a user to select whether or not to operate the backup means. . 16. The text-to-speech conversion system according to claim 1, wherein each means according to claim 5, each means according to claim 11,
A text-to-speech conversion system further comprising the means according to claim 14 and the means according to claim 15.