JPH0116160Y2 - - Google Patents

Description

[Detailed description of the invention] This invention relates to a sound generation device that inputs Roman letters.In particular, by operating the alphanumeric key and pressing a key representing a consonant and a key representing a vowel in combination, it is possible to input a single sound, or to input a vowel. Roman alphabet input that allows you to generate sounds while taking into account the meaning of the word or the spacing between words when generating sounds input in Roman alphabet format, such as inputting a single sound by pressing a key representing . relates to a sound generating device.

Prior to the filing of this application, the applicant provided an alphabet key, and by operating the keys representing consonants and the keys representing vowels of the alphabet key, or by operating the keys representing vowels, Japanese single sounds were written in Roman alphabet format. is input, the input single sound is generated and memorized at the same time, and based on the inputted word consisting of multiple sounds, a word representing a certain language (e.g. Japanese) corresponding to another language (e.g. English) is memorized in advance. We have proposed a so-called electronic dictionary that displays the translated word sounds read from memory and simultaneously generates them audibly. In this way, in a device that outputs input words in voice, the ability for the operator to program the voice can expand its uses. By the way, in Japanese, even if you read each character of the 50 syllables with pauses, the meaning of the word can be fully recognized. Digital data corresponding to single tones may be stored in a memory, and used to combine the respective single tones corresponding to key inputs to synthesize speech and output. For example, if you want to say "next time", press the alpha key □Z□I□K□A□
Enter I. However, in order to clarify the punctuation of each single note, use the space key SP and actually press □Z□ISP□K□ASP□ISP. and,
When you press the SP key, the previous single sound entered in the Roman alphabet is recognized, and "ZI" is "ji" and "KA".
is converted into the corresponding code "ka" and "I" is converted into the corresponding code "i", and this code is stored in the memory. When outputting, the code stored in the memory is read out and converted into the corresponding speech synthesis data. Generate a single note.

However, in the method of inputting single notes using the alphanumeric keys as described above, the space key must be pressed for each single note, which increases the number of key presses when inputting settings for single notes.
The drawback is that it takes a long time to input. Therefore, when inputting a single sound using an alpha key, if the pressed alpha key recognizes a "consonant + vowel" or "vowel" as a pair, the single sound can be input in Roman alphabet format without the need to press the space key. can do. Also, like "SHI", the 50-sound "S" is a 3-letter alphabet.
Even when inputting "consonant + consonant + vowel"
Therefore, it is only necessary to recognize that a single sound has been input when a key representing a vowel is pressed. Therefore, the present invention is based on "consonant + vowel" or "vowel"
Or the alphanumeric keys that represent vowels (i.e. A, I, U,
E, O) is input, the key representing the consonant input earlier (i.e. K, S, T, N, H,
A single note can be recognized as 50 notes by combinations of letters (M, Y, R, W, etc.). However, when a key representing a vowel is pressed, the 50 sounds cannot be recognized in combination with the key representing the consonant input just before, or only by the key representing the vowel, in the following cases: A situation may arise. For example, "N"
In other words, when entering a word that includes "N", if you enter "BUNAN", it will be translated as "BUNAN".
The problem arises in that it is not possible to distinguish between the two.

This invention was made to solve the above-mentioned problems, and it reduces the number of key operations and input time when inputting single sounds in the Roman alphabet using alphanumeric keys. To provide a speech generating device by inputting Roman characters, which allows inputting a single sound that has the meaning intended by a user when it includes "N" (N).

To summarize this idea, a special key (for example, haiphone key □-) is provided to indicate when a single vowel input by operating the alphanumeric keys cannot be distinguished, and when inputting the aforementioned "bunan", the alphanumeric key is used. Enter □B□U□N□A□N using the bet key,
When entering "Bunan", enter □B□U□N□-□A□N to change "N" to (n) or 50
The consonant "N" in the N line of the sound is entered in such a way that it can be identified, and the sound of the word can be generated vocally based on that input.

Examples will be described below with reference to the drawings.

FIG. 1 is an illustrative view showing the appearance of an example electronic dictionary to which the present invention is applied. In the figure, the electronic dictionary 10 is provided with an alpha key group 12 including keys 12A to 12Z representing A to Z on the operating section of a housing 11, as well as a key □-13 representing hiphone and a space key SP14. and,
Furthermore, a display 15 and a speaker 16 are arranged.

The illustration shows a case where keys for 26 letters of the alphabet A to Z are provided, but if you want to input Japanese 50 sounds and voiced sounds, use the alpha keys corresponding to the consonants in each line of the 50 sounds and voiced sounds. It may be only the key corresponding to the vowel.

FIG. 2 is a concrete block diagram of one embodiment of this invention. In the configuration, code conversion means 20
These include a key encoder 21 that generates a key code for each alphabet input from the alpha alphabet key group 12, a buffer register 22 that can store a key code corresponding to a three-character alpha alphabet, and an output of the buffer register 22. 50 key codes corresponding to each alpha key based on
A 50-tone code converter (described in detail in Figure 3 below) 23 that converts to a note code, and a hi-phone key 13
and an OR gate 24 that commands to write a 50-sound code based on either the space key 14 or the output of a vowel detection circuit 28, which is an example of vowel detection means.
and an input control circuit 2 that commands writing to memory.
5, a delay circuit 26, and a clear circuit 27 for clearing the buffer register. The vowel detecting means uses alpha keys (A,
Vowel detection that detects the presence or absence of a vowel input by determining whether any of I, U, E, O) is stored in the digit position corresponding to the last input key of the buffer register 22. This is the circuit 28. Note that the single-note alpha alphabet key code stored in the buffer register 22 included in the code conversion means 20 is given to the decoder 17 and displayed on the display 15.

The storage means 30 is a memory including a plurality of addresses (for example,
RAM) 31, and an address designation circuit 32 that designates a write or read address in the RAM 31.

The voice generating means 40 includes a gate circuit 41 that supplies the code read from the RAM 31 to the voice synthesis circuit 44 only when the read command signal S _v is applied.
, a space code discrimination circuit 42 that discriminates space codes among the codes output from the RAM 31, a delay circuit 43, a speech synthesis circuit 44, and a digital value that is correlated to the 50-note code synthesized by the speech synthesis circuit. a D/A conversion circuit 45 for converting into an analog value required for the output of the speaker 16;
It consists of:

FIG. 3 is a concrete block diagram of the 50-note converter 23. The outline of the 50-tone converter 23 will be described with reference to FIG. code converter 231
converts the input single-note key code given from the buffer register 22 into a 50-note code based on the constantly given 50-note (a, in...in) codes C _A , C _I ...C _N. Convert to gate circuit 23
Give to 7. This gate circuit 237 is input
If the 50-sound code is other than "N" (N), it will be output as is, and if it is "N", a vowel will be input after that, and it may be "N" representing the consonant in the N line.
The derivation of the 50 note code is stopped until there is an output from the AND gate 232 or 233. If the single sound to be input is "n" in the middle of a word, press the key 12N.
Since the hiphone key 13 is pressed when
The AND gate 232 supplies the 50-tone code C _N for "n" to the gate circuit 237 for output. In addition, if the final sound of the word is "N", and the key code stored in the first digit of the buffer register 22 is "N", the comparator 234 will output the preset "N". A high level representing a match with the key code (K _N ) of `` is derived and applied to the AND gate 233 . At this time, since the space key 14 is pressed when the key 12N is present, the AND gate 233 derives the code for "N" ( _CN ) and supplies it to the gate circuit 237 for output. after that,
Since the delay circuit 235 derives the high level after a certain time delay after pressing the space key 14,
The AND gate 236 derives the space code (C _sp ) and supplies it to the gate circuit 237 for output. Note that even if the final sound of the word is other than "n", the space code (C _sp ) is derived after a certain period of time (the time required to derive the 50-note code for a single note) after pressing the space key 14. .

Next, the concrete operation of this invention will be explained with reference to FIGS. 1 to 3. For example, assuming that the above-mentioned "Bunan" is to be input, the alphabet keys □B□U□N□A□N are pressed in sequence. now,
Key
When 12B is pressed, the key encoder 21 generates a key code corresponding to "B" and stores it in the first digit of the buffer register 24, and when the key 12U is subsequently pressed, the key code for "B" is generated. is stored in the second digit, and the key code "U" is stored in the first digit (as shown). At this time, the vowel detection circuit 28 determines that the key code of "U" representing the vowel "U" is stored in the first digit of the buffer register 22, and sends it to the input control circuit 25 via the OR gate 24. A write command is given to the delay circuit 26, and an operation command signal is given to the delay circuit 26. At this time,
The 50-note converter 23 derives a code representing the 50-note ``BU'' corresponding to ``BU'' based on the two-digit contents of the buffer register 22, and sends the code to the input control circuit 2.
5, the input control circuit 25 is
The code is sent to RAM via addressing circuit 32.
Write to the first address of 31. Thereafter, a signal is derived from the delay circuit 26 after a certain time delay determined by the delay circuit 26, and the clear circuit 27 clears the contents of each digit of the buffer register 22.

Next, press keys 12N and 12A to select "Na".
, the key code corresponding to "N" is stored in the second digit of the buffer register 22, and the key code corresponding to "A" is stored in the second digit of the buffer register 22.
The key code is stored in the first digit of the buffer register 22. Accordingly, the vowel detection circuit 2
8 is the vowel "A" representing "a", a high level signal is derived, and the OR gate 24
The signal is applied to the input control circuit 25 and the delay circuit 26 via the input control circuit 25 and the delay circuit 26. Therefore, the input control circuit 25 stores the "Na" code converted into a 50-note code by the 50-note converter 23 to the next address of the RAM 31 via the address circuit 32. After a certain time delay, the delay circuit 26 derives a high level and issues a clear command to the clear circuit 27, whereby the clear circuit 27 clears the contents of each digit of the buffer register 22.

Subsequently, when the key 12N representing "N" is pressed, the key encoder 21 stores the key code of N in the first digit of the buffer register 22. At this time, the vowel detection circuit 28 does not detect a vowel. Then, the user presses the space key 14 indicating that input of one word has been completed through the above operation. In response to pressing this space key 14,
A high level is given to the 50-tone converter 23, and is passed through the OR gate 24 to the input control circuit 25.
and the delay circuit 26. In response, the input control circuit 25 writes the "N" code (C _N ) converted by the Japanese 50-tone converter 23 to the subsequent address of the RAM 31 via the address circuit 32 as shown in FIG. After that, the 50-tone converter 23
Based on the presence of a high-level input from the space key 14, a space code (C _sp ) indicating that speech generation should be paused after the previously input word is generated and given to the input control circuit 25; The input control circuit 25 causes a space code to be written to a subsequent address in the RAM 31 via the address circuit 32.

Next, to explain the case where the sound of the word written in the RAM 31 is generated as described above, the gate opening command signal _Sv is given to the gate circuit 41, and the start address designation command signal Sv is given to the address circuit 32. _a is given. Therefore, when the address circuit 32 sequentially specifies addresses from the first address of the RAM 31 and commands to read out words stored in the RAM 31, the codes for each single note of the word stored in the RAM 31 are sequentially read out, and the gate circuit 41
The signal is applied to the speech synthesis circuit 44 via the . The speech synthesis circuit 44 converts the single sound read from the RAM 31 into
Digital information for generating 50 sounds based on the 50 sounds code is read out and given to the D/A conversion circuit 45. The D/A conversion circuit 45 converts the digital information read out from the voice synthesis circuit 44 to generate 50 tones of voice into an analog signal, and outputs the analog signal to the speaker 16.
to be generated as a single note. As a result, the speaker 16 utters "Bunan". At this time, the space code determination circuit 42 always
Since the output of 31 is determined whether or not it is a space code, when the space code (C _sp ) derived after generating the sound of the word "bunan" is detected, a delay signal is given to the delay circuit 43. , by prohibiting the address circuit 32 from specifying the address of the RAM 31 that stores the single sound of the next word for a predetermined period of time, the generation of sound is paused for a certain period of time. This has the advantage that word breaks can be easily understood because the generation of words is paused.

Next, a case where "Bunan" (draft) is inputted will be explained. To enter "draft", use alpha key group 12, □B□U□N□-□A□NSP
Enter Now, assuming that "BU" is input, the buffer register 22 has "U" in the first digit.
The key code "B" will be stored in the second digit. At this time, the vowel detection circuit 28 detects that "U" representing a vowel is stored in the first digit of the buffer register 22, and sends a high level to the input control circuit 25 and delay circuit 26 via the OR gate 24. give. In response, the input control circuit 25 causes the 50-note converter 23 to convert "BU" into the 50-note "BU" and writes the code to the address of the RAM 31 designated by the address circuit 32.

Next, when the key 12N representing "N" is pressed, the key code "N" is stored in the first digit of the buffer register 22, but at this time the vowel detection circuit 28 detects the vowel. It doesn't detect something. Thereafter, the user presses the hiphone key 13. In response to this pressing of the hi-phone key 13, a high level signal is applied to the Japanese 50 note converter 23, and a high level signal is applied to the input control circuit 25 and delay circuit 26 via the OR gate 24. Therefore, the input control circuit 25 has a 50-tone code corresponding to "N" coded by the 50-tone converter 23.
The code for the sound “n” is sent through the address circuit 32.
Write to the next address in RAM31. At this time,
Even if the hi-phone key 13 is pressed, there is no interval of one note. Then, after a certain time delay determined by the delay circuit 26, the clear circuit 27 clears each digit of the buffer register 22.

Further, when the key 12A representing "A" is pressed, the key code of "A" is stored in the buffer register 22. Therefore, the vowel detection circuit 28 detects that the vowel is stored in the first digit of the buffer register 22 and derives a high level output, which is sent to the input control circuit 25 and delay circuit 26 via the OR gate 24. give. As a result, the input control circuit 25 stores the code "A" in the RAM.
Write control is performed to the address following 31.

Next, the user presses the alphanumeric key 12N representing "n" and then presses the space key 14. As a result, the "N" code (C _N ) and the space code (C _sp ) are stored in the subsequent address of the RAM 31 in the same way as the above-described operation.

When reading out the audio of the word "draft" stored in the RAM 31 as described above, the address circuit 32 is designated with the first address and the gate 41 is commanded to open in the same manner as described above. "Boom-ang" is uttered.

If necessary, the contents stored in the buffer register 22 are decoded by the decoder 17 and displayed on the display 15 as an input alphanumeric value.

As described above, according to this invention, it is determined that a single sound has been input by detecting a vowel, and at the same time, if "n" appears in the middle of a word, it can be identified by using a specific key. Therefore, it is no longer necessary to use the space key for each single note input, simplifying the key input operation and shortening the key input operation time.Moreover, it is possible to reliably input the desired single note in Roman alphabet format without making any mistakes and record it by voice. This has the unique effect of being able to occur as

This invention can be used to simply generate sounds by inputting Roman characters, but it can also be effectively used in electronic dictionaries and the like when inputting Japanese single sounds in Roman characters.

[Brief explanation of the drawing]

FIG. 1 is an external view of an example to which this invention is applied. FIG. 2 is a block diagram of one embodiment of this invention. Figure 3 is a concrete block diagram of the 50-tone converter. In the figure, 12 is an alpha key, 13
14 is a haiphone key, 14 is a space key, 20 is a code conversion means, 30 is a storage means, and 40 is a voice generation means.

Claims (1)

[Claims for Utility Model Registration] Alphabetical keys for inputting a single sound to be generated in Roman alphabet format, a specific key for identifying a single sound that cannot be determined by inputting only with the alphanumeric keys, and operation of the alphanumeric keys. a vowel detecting means for detecting that an alphanumeric key representing a single vowel has been pressed by the alphanumeric key; Convert it to a code corresponding to a single note,
a code converting means for deriving a code corresponding to a single note inputted by the alpha-abbetical key according to the operation of the alpha-abbetical key and the specific key; 1. A sound generation device using Roman alphabet input, comprising: a storage means for storing each unit representing a sound; and a sound generation means for converting into sound and generating sound in response to reading of a code for each phoneme stored in the storage means.