JP2019204237A - Vocalization supporting system, input device, and conversion program - Google Patents

Abstract

To provide a vocalization supporting system, an input device, and a conversion program that can realize conversation without discomfort by making audio output at an atmosphere of natural conversation possible while making simple character input possible.SOLUTION: A vocalization supporting system comprises: a converter using an input device comprising eight input keys accepting an input operation with a left hand little finger, a left hand ring finger, a left hand middle finger, a left hand index finger, a right hand index finger, a right hand middle finger, a right hand ring finger, and a right hand little finger of a user and an enter key accepting an input operation with a left hand thumb and/or a right hand thumb of the user to specify a word corresponding to input information of a single sentence by the input device; and a speaker performing voice output of the word.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a speech support system that speaks on behalf of a user, and an input device and conversion program used in the speech support system.

  When a person with a disability who is unable to speak due to illness, etc., to communicate with others, he / she generally enters writing with paper and writing utensils, writing with a notebook computer, etc., or entering into a notebook computer, etc. In many cases, voice output (for example, refer to Patent Document 1) by reading out the sentence data is used.

JP 59-79337 A

  However, in writing using paper or writing using a notebook computer or the like, writing speed is slower than normal conversation speed, so it is difficult to perform sufficient communication, and stress is often felt. Moreover, in the case of writing, it has not been possible to apply to a communication method using only voice such as a telephone as a medium.

  For voice output by reading on a notebook computer etc., it is necessary to carry the notebook computer etc., and also a place for input operation to the notebook computer etc. is required, and the usable environment is limited. There was also a problem that.

  Recently, there is software that synthesizes speech using a so-called smartphone. However, in the smartphone input mode, it is necessary to input characters while staring at the keys displayed on the smartphone screen. It was inconvenient as a face-to-face conversation environment.

  Various special simple keyboards have been reported and proposals have been made to simplify input operations. However, training is necessary for smooth operation of all types of keyboards, and this is a burden on the user. Was a big one.

  In view of the current situation, the present inventor conducts research and development to realize a conversation without a sense of incongruity by enabling voice output in a natural conversation atmosphere while enabling easier character input, The present invention has been achieved.

  In the speech support system of the present invention, the user's left hand little finger, left hand ring finger, left hand middle finger, left hand index finger, right hand index finger, right hand middle finger, right hand ring finger, and right hand little finger An input device having an enter key for accepting an input operation of the thumb and / or right-hand thumb, a converter for specifying a word corresponding to single phrase input information by the input device, and a speaker for outputting the word by voice It is to be prepared.

In particular, in the input operation by the input device, the following rules are used.
・ Input operation of the first input key with the left hand little finger → Input of “Q”, “A”, “Z” ・ Input operation of the second input key with the left hand ring finger → Input of “W”, “S”, “X” 3rd input key input operation with middle finger → "E""D""C" input / 4th input key input operation with left index finger → "R""T""F""G""V" “B” input / input operation of fifth input key with right index finger → “Y”, “U”, “H”, “J”, “N”, “M” input / input operation of sixth input key with right hand middle finger → Input of “I” and “K” ・ Input operation of seventh input key with right ring finger → Input of “O” and “L” ・ Input operation of eighth input key with little finger of right hand → Input of “P”

Also,
The input operation of the eighth input key with the right little finger may be input as “P” and “−”.

  Furthermore, in the utterance support system and the conversion program of the present invention, when there is no word corresponding to the order information input by the input device, a connected word such as “um” is output, and further, a predetermined word at the input device is output. It is also characterized by outputting negative words such as “No” by performing an input operation.

  According to the utterance support system of the present invention, a simple input operation is performed by outputting a voice for each single phrase, and a smooth speech output is possible by the simplified input operation, thereby enabling natural conversation. be able to. In addition, the input device and the conversion system used in the utterance support system of the present invention can realize a conversation with less burden on the user and without feeling uncomfortable.

It is explanatory drawing of a keyboard of qwerty type. It is explanatory drawing of the speech support system which concerns on this invention. It is a flowchart of the conversion program which concerns on this invention. It is a flowchart of the conversion program of other embodiment.

  In the speech support system of the present invention, the user's left hand little finger, left hand ring finger, left hand middle finger, left hand index finger, right hand index finger, right hand middle finger, right hand ring finger, and right hand little finger An input device having an enter key for accepting an input operation of the thumb and / or right-hand thumb, a converter for specifying a word corresponding to single phrase input information by the input device, and a speaker for outputting the word by voice It is to be prepared.

  By outputting speech as a conversion for each single phrase, it is possible to realize a conversation with little discomfort without having the conversation partner wait for a long time.

  In addition, since the conversion is performed for each single phrase, even if the Roman input of the alphabet using the eight input keys is performed, it is rarely converted into an unexpected word in a general conversation.

Especially for input devices
・ Input operation of the first input key with the left hand little finger → Input of “Q”, “A”, “Z” ・ Input operation of the second input key with the left hand ring finger → Input of “W”, “S”, “X” 3rd input key input operation with middle finger → "E""D""C" input / 4th input key input operation with left index finger → "R""T""F""G""V" “B” input / input operation of fifth input key with right index finger → “Y”, “U”, “H”, “J”, “N”, “M” input / input operation of sixth input key with right hand middle finger → Input of “I” and “K” ・ Input operation of seventh input key with right hand ring finger → Input of “O” and “L” ・ Input operation of eighth input key with little finger of right hand → Input of input of “P” It is an operation. That is, as shown in FIG. 1, the rule of finger usage on a so-called general qwerty type keyboard is used, and if the blind touch of the qwerty type keyboard is possible even for a handicapped person, the voice of the present invention will be described. The support system can be used immediately, and the burden of training for use can be greatly reduced.

Also,
The input operation of the eighth input key with the right little finger may be input as “−” indicating a long sound as input of “P” and “−”.

Moreover,
・ Input operation of the sixth input key with the middle finger of the right hand → Input of “I”, “K”, “,” ・ Input operation of the seventh input key with the ring finger of the right hand → Input of “O”, “L”, “.” A comma or period may be input.

Hereinafter, the voice support system and the input device of the present invention will be described in detail.
As shown in FIG. 2, the utterance support system of the present invention includes a main body device including a converter (not shown) that specifies words corresponding to input information from the input device 10 and a speaker 21 that outputs the words as voice. 20 and an input device 10 connected to the main body device 20. In the present embodiment, the input device 10 and the main body device 20 are wired, but may be wirelessly connected.

  The input device 10 includes a first input key 11 that can be detachably attached to the fingertip of the user's left hand little finger, and a second input key 12 that can be detachably attached to the fingertip of the user's left hand finger. A third input key 13 detachably attachable to the fingertip of the middle finger of the user's left hand, a fourth input key 14 detachably attachable to the fingertip of the user's left index finger, A fifth input key 15 that can be detachably attached to the fingertip of the right index finger, a sixth input key 16 that can be detachably attached to the fingertip of the user's right middle finger, and the fingertip of the user's right hand ring finger A seventh input key 17 that can be detachably attached, an eighth input key 18 that can be detachably attached to the fingertip of the user's right hand, and a fingertip of the user's left thumb and right thumb It consists of an enter key 19 that can be detachably attached.

  Instead of attaching the enter key 19 to the left thumb and right thumb respectively, for example, the enter key 19 is attached to the left thumb, while the delete key (not shown) is attached to the right thumb. If you make a mistake, you can correct it with the delete key.

  The 1st to 8th input keys 11 to 18 and the enter key 19 are momentary switches attached to the fingertips of each finger, and an input operation by the momentary switch is performed by bringing the fingertips into contact with something. It will be. In the present embodiment, the hand with the first to eighth input keys 11 to 18 and the enter key 19 is placed on the user's own thigh, and the input operation is performed by touching the thigh with each fingertip. . The input operation is not limited to the thigh. For example, the input operation may be performed by touching an appropriate table, and the input operation can be performed by bringing a fingertip into contact with a familiar object.

  In the embodiment shown in FIG. 2, the first to eighth input keys 11 to 18 and the enter key 19 are attached to each fingertip by using double-sided tape or the like. By attaching the -8 input keys 11 to 18 and the enter key 19, respectively, it can be easily attached and detached. Alternatively, for example, an on / off switch may be provided at the position of each fingertip on the surface of an object having a predetermined shape that can be grasped with one hand, such as a ball.

  In the present embodiment, the main body device 20 is hung around the chest of the user via an appropriate strap 22, and the speaker 21 provided on the main body device 20 is pointed at the conversation partner.

  The main body device 20 accommodates a circuit board constituting a converter in a housing having a predetermined shape, and allows a predetermined sound output from a speaker 21 connected to the circuit board. Note that the output voice can be a voice close to the voice when the disabled person was able to speak.

  The circuit board constituting the converter is provided with an input terminal that receives an input signal from the input device 10, a memory that stores a conversion program to be described later, a microprocessor that executes the conversion program, and the like. A word corresponding to input information of a single phrase is specified.

In the memory, word correspondence data corresponding to the input information is stored in advance as a database. In particular, in this correspondence data, for example,
-Input operation of the first input key with the left hand little finger->"1" input-Input operation of the second input key with the left hand ring finger-> Input of "2"-Input operation of the third input key with the left middle finger->" 3 ”input / input operation of the fourth input key with the left index finger → input of“ 4 ”/ input operation of the fifth input key with the right hand index finger → input of“ 5 ”/ the sixth input key with the middle finger of the right hand Input operation → “6” input / input operation of seventh input key with right hand ring finger → input of “7” / input operation of eighth input key with right hand little finger → input of “8” In the converter, the input information input by the input device 10 is recognized as numeric string data.

  That is, for example, if you want to say "Today's head hurts", enter "Today", "Head" and "Pain" separated into single phrases, but the user can input "KYOHA" ", The converter recognizes that the numeric string data" 657551 "has been input. Then, the converter identifies “Today is”, which is a word registered in advance as the numeric string data, from the database, and outputs the speech data of “Today” from the speaker 21. The audio data output here may be pre-stored in the memory as audio data, or the audio data is synthesized by activating text data reading software and reading “Today” You may let them.

  When the user inputs “ATAMAGA (head)” as a Roman character input after voice output “Today is”, the converter recognizes that the character string data “1415141” is input, and this number string data When the user inputs “ITAI (pain)” as a Roman character input, the converter inputs character string data “6416”. It is recognized that the word “pain”, which is a word registered in advance as the numeric string data, is output as a voice.

  In this way, erroneous conversion can be prevented by previously registering the correspondence data of words corresponding to the input information in the database.

Incidentally, for example, in the case of numeric string data “6416”, 6 → I, K as a combination
4 → R, T, F, G, V, B
1 → Q, A, Z
6 → I, K
Therefore, it is considered that “IRAI” and “IGAI” can cope with other than “ITAI (pain)”, but in the case of single phrase input, “IRAI” “ There will be no clause only for “IGAI”, and there is no problem in registering the numeric string data “6416” as “ITAI (pain)” in the database.

  Also, the converter regards input of character string data for which the corresponding data is not registered in the database, and thinks that it was input incorrectly, and outputs speech data by associating connected words such as “E-to”. It is said. In this way, by outputting the connected words as a voice, the user is informed that the user has made an incorrect input and is prompted to re-input, while at the same time the conversation partner can be connected and feel uncomfortable. It can be made difficult.

  Furthermore, if the voice output word is different from the user's expected word, the enter key 19 is continuously input by the user's left thumb and / or right thumb input operation. In the converter, negative words such as “No” are output as voice. In this way, by outputting a negative word immediately in response to an erroneous output, it is possible to eliminate the possibility of causing erroneous recognition in the conversation partner.

  Below, based on the flowchart of the conversion program shown in FIG. 3, it demonstrates in full detail about the conversion process in a converter.

  First, when the conversion program is activated, it enters a standby state for an input operation from the input device 10, and waits for input of the first to eighth input keys 11 to 18 (step S1).

  When there is no input from the first to eighth input keys 11 to 18 in step S1 (step S1: NO) and there is no input from the enter key 19 (step S2: NO), the conversion program The standby state is maintained for the input of the 1 to 8 input keys 11 to 18.

  In the conversion program, the first to eighth input keys 11 to 18 are input in step S1 (step S1: YES). When the enter key 19 is input (step S3), the first to eighth input keys 11 to 11 are input. The number string data input by 18 is specified, the database is searched, and the correspondence data of words corresponding to the number string data is specified (step S4).

  When there is word correspondence data corresponding to the number string data in step S4 (step S4: YES), the conversion program causes the speaker 21 to output the words of the correspondence data by voice (step S5). After step S5, the conversion program returns to step S1 and waits for the next input of the first to eighth input keys 11-18.

  On the other hand, if there is no word correspondence data corresponding to the numeric string data in step S4 (step S4: NO), the conversion program outputs a connected word such as “Et” (step S6). ). After step S6, the conversion program returns to step S1 and waits for the next input of the first to eighth input keys 11-18.

  On the other hand, if the enter key 19 is input in step S2 (step S2: YES), the enter key 19 is continuously input in step S2 after the enter key 19 is input in step S3. Therefore, a negative word such as “No” is output by voice (step S7). After step S7, the conversion program returns to step S1 and waits for the next input of the first to eighth input keys 11-18.

  In this way, in the conversion program, when a voice output unintended by the user is performed or when a voice output unintended by the user is performed, a negative word or a connection word is automatically output, It is possible to produce a natural conversation and make it difficult for the conversation partner to feel uncomfortable.

  As another embodiment, when a plurality of correspondence data must be registered for one input information in a database in which correspondence data of words corresponding to the input information is registered in advance, the conversion shown in the flowchart of FIG. It can also be a program.

  That is, when this conversion program is started, it enters a standby state for input operations from the input device 10, and waits for input of the first to eighth input keys 11 to 18 (step S1 ').

  In the conversion program, the first to eighth input keys 11 to 18 are input in step S1 ′ (step S1 ′: YES). When the enter key 19 is input (step S2 ′), the first to eighth inputs are input. The number string data input by the keys 11 to 18 is specified, the database is searched, and the word correspondence data corresponding to the number string data is specified (step S3 ′).

  If there is word correspondence data corresponding to the numeric string data in step S3 ′ (step S3 ′: YES), the conversion program causes the speaker 21 to output the words of the correspondence data by voice (step S4 ′). .

  If the voice output word is not the word intended by the user, the user inputs the enter key 19 (step S5 ′: YES), and after the enter key 19 is input in step S2 ′. Since the input of the enter key 19 in step S5 ′ is continuously performed, the conversion program outputs a negative word such as “No” as a voice (step S6 ′).

  Further, the conversion program performs setting for designating the next candidate of the word correspondence data corresponding to the numeric string data (step S7 '). That is, a candidate designation parameter is provided, and normally the first candidate is selected, and when it is not the first candidate, the second candidate is selected.

  The conversion program performs the setting for designating the next candidate for the corresponding data in step S7 ′, and then returns to step S3 ′ to identify the second candidate corresponding data and the corresponding data exists (step S3 ′: YES), the words of the corresponding data are output as voice from the speaker 21 (step S4 ′).

  If the conversion program is not the second candidate, the conversion program causes the user to input the enter key 19 (step S5 ′: YES) and outputs a negative word such as “No” as a voice (step S6). '), The third candidate is selected in step S7'.

For example, in the case of the above-described number string data “6416”, three cases of “ITAI (pain)”, “IRAI” and “IGAI” will be described. Here, it is assumed that “ITAI (pain)” as the first candidate, “IRAI” as the second candidate, and “IGAI” as the third candidate are registered, and “IGAI ( If you want to output “Igai” ”as an audio output, the order of audio output is“ ITAI (pain) ”→“ No ”→“ IRAI ”→“ No ”→“ IGAI ”
Thus, the target “IGAI” can be output as a voice.

  If there is no word correspondence data corresponding to the number string data in step S3 ′ (step S3 ′: NO), the conversion program causes the connected words such as “Et” to be output as voice (step S8 ′), the process returns to step S1 ′, and the next input of the first to eighth input keys 11-18 is awaited.

  In this way, in the conversion program, when a voice output unintended by the user is performed or when a voice output unintended by the user is performed, a negative word or a connection word is automatically output, It is possible to produce a natural conversation and make it difficult for the conversation partner to feel uncomfortable.

  In the conversion program of the present invention, the on / off switch of the main device 20 is forcibly terminated when it is turned off.

10 input devices
11 First input key
12 Second input key
13 Third input key
14 Fourth input key
15 Fifth input key
16 Sixth input key
17 Seventh input key
18 Eighth input key
19 Enter key
20 Device
21 Speaker
22 Strap

Claims (12)

Eight input keys for receiving input operations by the user's left hand little finger, left hand ring finger, left hand middle finger, left hand index finger, right hand index finger, right hand middle finger, right hand ring finger, right hand little finger, and the left thumb and / or right thumb of the user An input device having an enter key for receiving an input operation;
A converter for identifying words corresponding to input information of a single phrase by the input device;
An utterance support system comprising a speaker for outputting the words as a voice. It is assumed that any one of alphabets “Q”, “A”, and “Z” is input by the input operation with the left little finger,
It is assumed that any of alphabets “W”, “S”, and “X” is input by the input operation with the left ring finger,
It is assumed that one of alphabets “E”, “D”, and “C” is input by the input operation with the middle finger of the left hand,
It is assumed that any one of alphabets “R”, “T”, “F”, “G”, “V”, and “B” is input by the input operation with the left index finger,
It is assumed that any of alphabets “Y”, “U”, “H”, “J”, “N”, and “M” is input by the input operation with the right index finger,
It is assumed that either “I” or “K” of the alphabet is input by the input operation with the middle finger of the right hand,
It is assumed that one of alphabets “O” and “L” is input by the input operation with the right ring finger,
The speech support system according to claim 1, wherein the alphabet “P” is input by an input operation with the right hand little finger.   The utterance assistance system according to claim 2, wherein either an alphabet “P” or “−” indicating a long sound is input by an input operation with the right hand little finger.   The utterance support system according to any one of claims 1 to 3, wherein the converter outputs a connection word such as "Uto" from the speaker when a word corresponding to the input information does not exist.   When the sound output from the speaker is different from the sound expected by the user, a negative input such as “No” is made by performing a predetermined input operation on the input device. The utterance assistance system according to any one of claims 1 to 4, wherein the utterance support system is output from the voicing system. A first input key in which any one of alphabets “Q”, “A”, and “Z” is input by an input operation with the left little finger of the user;
A second input key in which any one of alphabets “W”, “S”, and “X” is input by an input operation using the left ring finger of the user;
A third input key in which any one of alphabets “E”, “D”, and “C” is input by an input operation with the middle finger of the user's left hand;
A fourth input key in which any one of alphabets “R”, “T”, “F”, “G”, “V”, and “B” is input by an input operation using the left index finger of the user;
A fifth input key in which any one of alphabets “Y”, “U”, “H”, “J”, “N”, and “M” is input by an input operation with the right index finger of the user;
A sixth input key in which any one of alphabets “I” and “K” is input by an input operation with the middle finger of the user's right hand;
A seventh input key in which any one of alphabets “O” and “L” is input by an input operation with the user's right-hand ring finger;
An eighth input key on which the alphabet “P” is input by an input operation with the user's right little finger;
An input device comprising an enter key that is input by an input operation using the left thumb and / or right thumb of the user.   The input device according to claim 5, wherein either an alphabet “P” or “−” indicating a long sound is input by an input operation by the user's little finger. The first input key is detachably attachable to the fingertip of the user's left little finger,
The second input key can be detachably attached to the fingertip of the left ring finger of the user,
The third input key is detachably attachable to the fingertip of the middle finger of the user's left hand,
The fourth input key can be detachably attached to the fingertip of the left index finger of the user,
The fifth input key is detachably attachable to the fingertip of the user's right index finger,
The sixth input key is detachably attachable to the fingertip of the middle finger of the user's right hand,
The seventh input key is detachably attachable to the fingertip of the user's right hand ring finger,
The eighth input key is detachably attachable to a fingertip of the user's right little finger,
The input device according to claim 6 or 7, wherein the enter key is detachably attachable to a fingertip of a left thumb and / or a right thumb of the user. In the conversion program of the utterance support system that identifies a single-phrase word corresponding to the input information input by the user's input operation and outputs the word as sound from a speaker,
It is assumed that any one of alphabets “Q”, “A”, and “Z” is input by an input operation with the left little finger of the user,
It is assumed that any one of alphabets “W”, “S”, and “X” is input by an input operation with the left user's ring finger of the previous user,
It is assumed that any one of alphabets “E”, “D”, and “C” is input by an input operation with the middle finger of the user's left hand,
It is assumed that any one of alphabets “R”, “T”, “F”, “G”, “V”, and “B” is input by an input operation with the left index finger of the user,
It is assumed that any of alphabets “Y”, “U”, “H”, “J”, “N”, and “M” is input by the input operation with the user's right index finger,
It is assumed that one of alphabets “I” and “K” is input by the input operation with the middle finger of the user's right hand,
It is assumed that one of alphabets “O” and “L” is input by an input operation using the user's right hand ring finger,
Assuming that the alphabet “P” is input by the input operation with the little finger of the user's right hand,
A conversion program specifying the language.   The conversion program according to claim 9, wherein either “P” of alphabet or “−” indicating a long sound is input by an input operation with the little finger of the right hand of the user. The correspondence data of the word corresponding to the input information,
The conversion program according to claim 9 or 10, wherein a connection word such as "Uto" is associated with the input information that does not exist in the correspondence data.   Assume that an enter key is input by an input operation using the left thumb and / or right thumb of the user. When the enter key is continuously input, a negative word such as “No” is associated. The conversion program of any one of Claims 9-11.