JP2000315205A - Voice translating machine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a small and easy-to-hold voice translating machine by preparing its casing consisting of a grip part which is held by a single hand of a user and an information input/output part which includes a voice input means, a voice output means and an information display means. SOLUTION: A casing 100 consists of an information input/output part 150 including an information display means 11, a voice input means 1 and a voice output means and an operating grip part 151 which is held by a user. Furthermore, the casing 100 has a small size to be easily stored into the breast pocket of an overseas traveler, for example. The part 151 has a shape that can be gripped by a thumb and another finger of the user so that the user can operate a voice input switch 101 while holding the part 151 by his/her single hand. Then the user utters his/her desired words to be translated and inputs them. These inputted words are converted into their corresponding character strings and the candidate character strings are simultaneously displayed on the means 11 in the order of higher likelihood.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an interpreter for converting a voice uttered by a user into another language and outputting the converted voice.

[0002]

2. Description of the Related Art In the prior art, for example,
When you start the program on a personal computer like the Japanese-English simultaneous interpretation software of NEC Corporation listed in the morning edition of the Yomiuri Shimbun on the 16th of January, and speak Japanese into the connected microphone, it will be translated into English and spoken There is an example to output with. Japanese Patent Application Laid-Open No. Hei 8-265445 discloses an electronic translator for translating an intermediate language once and then translating it into another language.

[0003]

As described above, in the Japanese-English simultaneous interpretation software described in the prior art, if the user wants to have simultaneous interpretation, it is general-purpose personal computer software. However, it is necessary to carry a personal computer, and when translating, it is necessary to start up the OS and then start up the software. In an actual use situation such as overseas travel, it is not always easy to use the system. Was.

Japanese Patent Application Laid-Open No. Hei 8-265445 discloses an electronic translator for a wireless telephone connected to a wireless, wired or infrared network, and is a communication device in the form of a telephone handset. Each of them has one communication device, and when communicating, each device is used like a telephone, and a speaker and a microphone are placed near your mouth and ears, so that the voice to be surely translated It was designed to capture audio and output audio.

[0005] On the other hand, the present invention assumes a case where one small interpreter held by one person communicates with a plurality of people facing each other in different languages.

An object of the present invention is to provide a voice interpreting apparatus that allows users in different languages to communicate with each other easily on the spot without being aware of the language and without choosing a time or place.

Another object of the present invention is to provide a small-sized interpreter which is easy to hold in view of the fact that a human being holds the small-sized interpreter of the present invention so as to be disposed for a long period of time with a person who intends to communicate. It is in.

Another object of the present invention is to provide a small-sized interpreter which can easily operate the interpreter operation with one hand.

Another object of the present invention is to provide a small-sized interpreter having a configuration that allows the other party to easily hear another interpreted natural language.

[0010]

According to the present invention, there is provided a function of converting a language in a language input by voice through a built-in microphone into a corresponding language in another language, and outputting the conversion result as voice from a speaker. The housing is small and has a shape that can be operated with one hand. Further, a speaker for outputting the sound is installed on a surface opposite to a surface on which the microphone is installed. The power switch is
Install it on a different part of the device from the grip.
In addition, the center of gravity of the entire apparatus is positioned at the grip portion. Further, the volume control switch is arranged so as not to be operable when the housing is placed on a flat surface. Further, the voice input means is provided above a portion different from the grip portion. Also, the audio output means is installed in a recessed portion of the housing.

[0011] To achieve still another object, the present invention provides:
The first natural language voice is input to a housing that internally includes a processing device that recognizes the input first natural language voice, converts the recognized content into a second natural language, and outputs the voice. The microphone and the speaker that outputs the sound of the second natural language are provided on different sides. Preferably, the microphone and the speaker are provided on opposite sides. With this configuration, when the present invention is used by arranging it between a plurality of persons, the first natural language voice to be interpreted is blocked by the small interpreter of the present invention, and thus the second natural language voice to be converted and output is output. The second natural language is not disturbed by the first natural language voice, but is output in a manner that is easy to hear to the other party, and communication can be taken well.

[0012] To achieve still another object, the present invention provides:
The small interpreter's housing has a grip that can be grasped with the palm of the hand, and the grip is rounded. Thereby, even if the interpreting operation is performed by lifting the small interpreting machine for a long time, the degree of fatigue can be reduced. Furthermore, since the receiving of the housing is arranged on the grip portion, the holding becomes easier. In particular, if a heavy battery storage portion is preferably disposed in the grip portion, stable holding for a long time is possible. When a lightweight battery is employed, another structure inside the housing may be arranged inside the grip.

Further, since the operation button is provided in the moving range of the thumb of the grip portion, it is possible to smoothly perform the processing operation of the voice interpretation without disturbing the communication while keeping the positions of the microphone and the speaker in a stable position. it can.

[0014]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram of a Japanese-English speech interpreter showing one embodiment of the present invention. This embodiment is a Japanese-English audio interpreter that translates Japanese into English, but this is merely an example. It translates English into Japanese (English and Japanese) and translates Japanese into Chinese (Japanese and Chinese). ), And the type of language is not limited, such as translating Chinese into Japanese (Chunichi). In the figure, 1 is a voice input unit, 2 is a voice recognition unit, 3 is a central processing unit, 4 is a language translation unit, 5 is a voice output unit, 6 is a memory, 7 is a voice generation unit, 8 is a unit set, 9 is a language conversion table, 10 is a command input switch, and 11 is information display means. FIG. 2 shows the data structure of the memory 6. FIG. 3 is an external view of the Japanese-English speech interpreter, which shows one embodiment of the present invention, as viewed obliquely from the front. In the following, an example of a device in which the dominant hand is intended for a right-handed user will be described, but when the left hand is intended for a dominant-handed user, a device in which the left and right sides are reversed is configured. I just need. In FIG. 1, reference numeral 100 denotes a housing, and the housing 100 includes an information input / output unit 150 on which the information display unit 11, the audio input unit 1, and the audio output unit 5 are mounted, and an operation grip unit 151 held by a user. Reference numeral 101 denotes a voice input switch that enables the voice input unit 1 to input voice, and reference numeral 102 denotes a power switch. Further, FIG. 4 is an external view of the Japanese-English speech interpreter, which shows one embodiment of the present invention, as viewed obliquely from behind. 3, as in FIG. 3, reference numeral 100 denotes a housing, and the housing 100 is provided with a display unit 150 equipped with the information display unit 11, the voice input unit 1, and the voice output unit 5 and an operation grip unit 151 held by a user. Become. 107 is a volume control switch for controlling the volume of the audio output from the audio output means 5; 108 is a determination switch for determining if the result displayed on the information display means 11 is correct; A cancel switch for returning to the state, 110 is an audio output switch for outputting audio from the audio output means 5, and 105 is an information display means 11.
Is an up switch for moving the cursor displayed on the information display unit 11 upward, and 106 is a down switch for moving the cursor displayed on the information display unit 11 downward. The command input switch 10 in FIG. 1 includes a voice input switch 101, a volume control switch 107, an enter switch 108, and a cancel switch 10.
9. The audio output switch 110, the up movement switch 105, and the down movement switch 106.

The device of the present invention has a housing sized to be stored in a breast pocket of clothes so that a user such as an overseas traveler can easily carry the device. The housing size will be described with reference to FIG. 5, FIG. 6, FIG. 7, and FIG. As shown in FIG. 5, the height of the main body 100 is a, the width of the operation grip part 151 is c, the thickness is e, the width of the information input / output unit 150 is d, and the thickness is b. In addition, the width of the breast pocket 120 of clothing such as suits is p and the height is q. At this time, a + 2 × b <pd <q. For example, in the case of an M size Japanese male suit, 80 mm <p <90 mm, 100 mm <q <120
Since it is about mm, the smallest one is adopted assuming the user, and a + 2 × b <80 d <100. Of course, the values of p and q in the case of the Japanese male of the M size are merely examples, and the width and height of the breast pocket of the clothes of the target user using the apparatus of the present invention may be used. The width and height may be an average length. By doing so, as shown in FIG. 8, the user can store the apparatus of the present invention in a breast pocket of clothing such as a suit, and can easily carry it when traveling overseas.

Further, the device of the present invention is formed in such a shape that the operation grip portion 151 can be gripped so as to be sandwiched between the thumb and other fingers during use, as shown in FIG. Therefore, as shown in FIG. 7, if the length from the base to the tip of the index finger of the human hand is s, and the length from the base to the tip of the thumb is t, then e + 2 × c ≒ s + t. For example, in the case of a Japanese adult male, since s = 120 mm and t = 60 mm, e + 2 × c ≒ 180 mm assuming the user. Of course, s, t for this Japanese adult male
Is merely an example, and can be replaced with the length of the finger of the target user using the device of the present invention. By doing so, the user can stably hold the device of the present invention with one hand.

Further, the shape of the apparatus of the present invention will be described with reference to FIGS. As described above, in the present invention, the operation grip portion 151 is shaped so as to be gripped so as to be sandwiched between the thumb and other fingers. Therefore, as shown in FIG. 9, four fingers other than the thumb are located on the front surface of the operation grip portion 151 and support the operation grip portion 151.
Therefore, in the present invention, the front surface of the operation grip portion 151 is curved in the vertical and horizontal directions. Also,
A rough material is used as the surface material of the operation grip portion 151, or the surface is processed to have irregularities. This allows the user to hold the device securely with only one hand. Also, as shown in FIG. 8, the device of the present invention is wound in the chest pocket of the clothing by bending the front surface of the operation grip portion 151 in the vertical direction so as to damage the clothing and the body. Without removing the device from the breast pocket of the garment.

Further, in the present invention, the operation grip portion 151 is provided.
Is set so that the center of gravity 300 of the entire apparatus is located within the range.
For this purpose, for example, a method may be used in which the power supply 302 such as a battery is configured to be set in the operation grip portion 151. Of course, the method of positioning the center of gravity 300 of the entire device within the range of the operation grip portion 151 includes:
Not only the method of setting the power supply to the operation grip section 151,
In addition, various methods can be used, such as using a heavy material or increasing the thickness of the material.
In this way, the user can securely hold the apparatus with only one hand without perceiving a feeling of weight.

Next, the operation and operation will be described.

In FIG. 3, first, when the power is turned on using the power switch 102, the voice input switch 101 of the command input switches 10 becomes operable. Here, when the user presses the voice input switch 101, the voice input means 1 is activated, and the voice input means 1 is enabled. In the present invention, as shown in FIG. 6, the power switch 102 is installed at a position where a finger holding the operation grip 151 cannot reach, that is, at the information input / output unit 150. In this way, it is possible to prevent an erroneous operation in which the power switch 101 is turned off by mistake while operating the apparatus. Also,
By doing so, as shown in FIG.
When the device is placed so that the power switch 101 is located at the bottom of the power switch 160, it is possible to prevent an erroneous operation in which the power switch 101 is turned on or off by mistake.

On the other hand, in the present invention, the audio input switch 101
Can be operated while the user holds the operation grip section 151 with one hand. For this purpose, as shown in FIG. 14, it can be pressed with the index finger. In that case, the voice input switch 101 is set so that its center is located at a distance V from the right end of the information input / output unit 150 in FIG. Note that v <u. u is the distance from the tip of the index finger to the second joint, as shown in FIG. 7, and in the case of a Japanese adult male,
Since u is approximately 50 mm, v <50 mm when the user is assumed. Of course, the value of u in the case of a Japanese adult male is merely an example, and the distance from the tip of the index finger of the target user using the apparatus of the present invention to the second joint may be used. By doing so, the voice input operation can be performed with only one hand.

Next, the user holds the apparatus with his right hand and uses the voice input means 1 to utter and input a word to be interpreted, for example, "Good morning". In the present invention, as shown in FIG. 4, the voice input unit 1 is installed at a position where a finger holding the operation grip unit 151 cannot reach, that is, at the rear surface of the information input / output unit 150. By doing so, the operation grip section
It is possible to prevent the finger to be held by covering the voice input means 1 from masking the input voice and to prevent the recognition accuracy of the input voice from lowering. In the present invention, the voice input means 1
Is installed in the upper part of the rear surface of the information input / output unit 150 as shown in FIG. By arranging the voice input means 1 on a different surface from the voice input switch 101 as described above, noise generated when the voice input switch 101 is pressed can be prevented from being mixed into the voice input means 1, and the recognition can be performed. Can be reduced. In addition, by arranging the voice input means 1 above the information input / output unit 150 in this way, it is possible to have a conversation while looking at the other person's face, and to facilitate communication.

Subsequently, the voice recognition means 2 is activated. The voice recognition means 2 first converts the input voice from analog to digital, and stores it in the WAVE 501 on the memory 6. The sampling rate at that time is 8kHz, 11kHz, 16kHz
Etc. can be appropriately determined.

Next, the voice recognition means 2
The voice data stored in E501 is read, the voice data is recognized, and converted into a character string corresponding to the voice. As a method of recognizing voice and converting it to a character string, for example, L. Rab
iner & B.-H. Juang, edited by Furui, "Basics of Speech Recognition (2)" (NTT Advanced Technology, 1995) p.245
To P304 can be used. Of course, other voice recognition methods may be used, and the voice recognition method does not limit the present invention. By this step, the voices are sorted in the order of likelihood, for example, by the character strings “refill”, “good morning”, “return”,
It is converted like "Ozuke". Subsequently, the RECOG (1) 50 on the memory 6 is sorted in order of likelihood from these candidate character strings.
5. Store in RECOG (2),…, RECOG (N). Here, for example, N = 20, but N can be determined as appropriate. Next, the voice recognition means 2 performs RECOG (1) 505, RE
COG (2), RECOG (3), and RECOG (4) are sequentially and simultaneously displayed on the information display means 11 as shown in FIG. At this time, "1" is stored in the ICNT 504 on the memory 6 so that RECOG (1) displayed on the information display means 11 is inverted between black and white. In the present invention, as shown in FIG.
11 is set at a position where a finger holding the operation grip section 151 cannot reach, that is, at the rear of the information input / output section 150. By doing so, it is possible to prevent the finger gripping the operation grip portion 151 from covering the information display means 11 to mask the displayed characters, and the speaker can easily check the characters.

At this point, if the user wants to cancel the input voice and re-enter the voice, the user can depress the cancel switch 109 so that the voice can be input again. When the cancel switch 109 is pressed, the cancel switch 109 is reset to the same initial state as when the power switch 102 is pressed, so that the voice input switch 101 can be operated.

As shown in FIG. 14, an operation grip portion 151 is provided.
It is installed within the range of the sector of the radius w and the angle θ from the lower right of the rear surface. For example, in the case of a Japanese adult male, w is approximately 80 mm and θ = 70 °. Therefore, assuming the user, the decision switch 108 has a sector shape with a radius of 80 mm and an angle of 70 ° from the lower right of the operation grip unit rear surface 151. Install within the range. Of course, the values of w and θ in the case of a Japanese adult male are merely examples, and the length of the thumb of the target user using the device of the present invention and the angle between the thumb and the index finger may be used. In this way, the operation can be canceled with only one hand.

If the display of the first candidate is correct, the user
The confirmation switch 108 is pressed as it is. Confirm switch 108
Also, as shown in FIG. 14, the operation grip portion is disposed within a sectoral shape having a radius w and an angle θ from the lower right of the rear surface 151 of the operation grip portion. For example, as mentioned above, in the case of a Japanese adult male,
Since w is approximately 80 mm and θ = 70 °, assuming the user, the determination switch 108 is connected to the operation grip unit rear surface 151.
It should be installed within a fan-shaped area with a radius of 80 mm and an angle of 70 ° from the lower right. Of course, w for this Japanese adult boy
The values of θ and θ are merely examples, and the length of the thumb of the target user using the device of the present invention and the angle between the thumb and the index finger may be used. By doing so, the candidate confirmation operation can be performed with only one hand. If the display of the first candidate is not correct, the user presses the down movement switch 106 to move the black-and-white inverted display area downward to a portion on the information display unit 11 where a correct character string is displayed. That is, every time the down movement switch 106 is pressed,
The value of the ICNT 504 on the memory 6 is incremented, and only the portion of the information display means 11 where the contents of the RECOG (ICNT) on the memory 6 are displayed is displayed in black and white inverted.
When the value of ICNT 504 becomes “M” (M ≧ 5),
COG (M-3), RECOG (M-2), RECOG (M-1), RECO
G (M) is displayed on the information display means 11.

If it has gone too far, the up / down switch 105 may be used to move the black-and-white inverted display area upward.
In other words, each time the upward movement switch 105 is pressed, the value of the ICNT 504 on the memory 6 is decremented, and only the portion of the information display means 11 on which the RECOG (ICNT) content on the memory 6 is displayed is inverted between black and white. To be displayed. As shown in FIG. 14, the lower switch 106 and the upper switch 105 are also installed within a sector of a radius w and an angle θ from the lower right of the rear surface 151 of the operation grip. As described above, for example, for a Japanese adult male, w is approximately 80
mm, θ = 70 °, so assuming the user, the lower movement switch 106 and the upper movement switch 105 are installed within a sector shape having a radius of 80 mm and an angle of 70 ° from the lower right of the rear surface 151 of the operation grip unit. To do. Of course, the values of w and θ in the case of a Japanese adult male are merely examples, and the length of the thumb of the target user using the device of the present invention and the angle between the thumb and the index finger may be used. By doing so, the candidate selection operation can be performed with only one hand. The user
When the correct character string among the character string candidates displayed on the information display means 11 is inverted between black and white, the confirmation switch 108 is pressed. In the example of FIG. 11, “Good morning” is displayed on the second line 402, so that the black-and-white inversion area is moved to the row 402 by pressing the down movement switch 106 and the confirmation switch 108 is pressed. . When the confirm switch 108 is pressed, the character string stored in the RECOG (ICNT) on the memory 6 is stored in the JAPANESE 502 on the memory 6 as well. In the example of FIG. 11, the selected candidate “good morning” is stored in the JAPANESE 502 on the memory 6.

Next, the language translating means 4 is activated. The language translator 4 uses the language conversion table 9 to store the JA on the memory.
Translates the character string stored in PANESE 502 into another language. Hereinafter, the operation of the language translating means 4 will be described.
FIG. 13 shows the data structure of the language conversion table 9.

First, the language translating means 4 executes the JAPA on the memory 6
The data “good morning” stored in the NESE 502 is sequentially collated with each item of the word 211 of the language conversion table 9, and the contents of the translation 212 of the matched item are stored in the ENGLISH 503 of the memory 6. In the above example, since the item 203 corresponds to the good morning, the corresponding translated word “good morning” 212 is stored in the memory 6.
Store in ENGLISH 503. For the conversion to the translated word, not only the method using the language conversion table as in the example described above, for example, a translation method using an example database described in JP-A-03-328585, and JP-A-03-51022 A translation method based on the described syntax analysis may be used.

Next, the language translating means 4 displays the contents stored in the JAPANESE 502 and the ENGLISH 503 of the memory 6 on the information display means 11, as shown in FIG. The display method shown in the figure is an example, and the display method does not limit the scope of the present invention.

Subsequently, the voice generating means 8 is activated. The voice generation means 7 reads the character string stored in the ENGLISH 503 on the memory 6, converts the character string into a synthesized voice, and stores the waveform data in the SYNWAVE 504 on the memory 6. To convert a character string into synthesized speech, for example, J. Allen, MS
Hunnicutt and D.Klatt et al. “From Text to Speech” (Cam
bridge University Press, 1987) Formant synthesis method described on pages 16 to 150 and arrow head "The latest status of text-to-speech synthesis technology" (Interface, Dec., 1996), page 161
And the waveform superposition method described on page 165 can be used. Of course, other text-to-speech synthesis methods can be used. Alternatively, a voice corresponding to each English translation may be recorded in advance and stored in a storage device such as a memory.

Here, when the user presses the voice output switch 110, the voice generating means 7 causes the SYNWAVE 50
The audio waveform data stored in 4 is converted into analog data by digital / analog conversion and output as audio through the audio output means 5. The audio output switch 110 also
As shown in FIG. 14, the operation grip unit 151 is installed within a fan-shaped range having a radius w and an angle θ from the lower right of the rear surface 151.
For example, as described above, in the case of a Japanese adult male, w is approximately 80 mm and θ = 70 °. Therefore, assuming the user, the voice output switch 110 is connected to the operation grip unit rear surface 151.
It should be installed within a fan-shaped area with a radius of 80 mm and an angle of 70 ° from the lower right. Of course, w for this Japanese adult boy
The values of θ and θ are merely examples, and the length of the thumb of the target user using the device of the present invention and the angle between the thumb and the index finger may be used. By doing so, the voice output operation can be performed with only one hand.

When it is desired to change the volume of the output sound, the volume can be adjusted by operating the volume control switch 107. The volume control switch 107 may be a rotary switch as shown in FIG. 14 or a slider switch. As shown in FIG. 14, the volume control switch 107 is also
It should be installed within the sector of radius w and angle θ from the lower right. For example, as described above, in the case of a Japanese adult male, w is approximately 80 mm and θ = 70 °, so that assuming the user, the volume control switch 107 has a radius of 80 mm from the lower right of the operation grip unit rear surface 151. , Within the range of a 70 ° angle fan. In this way, the volume control operation can be performed with only one hand. As shown in FIG. 15, a step having a depth f is provided between the information input / output unit 150 and the operation grip unit 151, and the volume control switch 107 is set within the sector shape and the information input / output unit 150 and the operation It may be provided near the step between the grip portions 151. The value of f may be a value that slightly exceeds the portion of the volume control switch 107 that is protruded from the rear surface of the operation grip portion 151.
Of course, the value of f cannot exceed the value of b. By doing so, when the device is placed on the flat surface 106 as shown in FIG. 15, the volume control switch 107 operates and the output sound suddenly becomes loud or cannot be heard. Can be prevented.

In the present invention, as shown in FIGS. 3 and 16, the audio output means 5 is installed on the opposite side to the installation surface of the audio input means 1, that is, on the side opposite to the installation side.
By doing so, it is possible to prevent a howling phenomenon in which the sound output from the sound output means 5 is input to the sound input means 1 and falls into an amplification loop. In addition, it is possible to proceed with the conversation without losing the posture or changing the way of holding the device when the dialogue is faced to each other. Further, since the device is located at the mouth of the speaker, the utterance is interrupted, and only the translated and output voice can be heard by the receiver.

In the present invention, as shown in FIGS. 3 and 6, the voice output means 5 is installed at a position where a finger holding the operation grip 151 cannot reach, that is, at the front of the information input / output unit 150. By doing so, the operation grip portion 15
The finger holding the cover 1 covers the sound output means 5 to prevent the output sound from being masked, so that the listener can easily hear the sound. Further, in the present invention, as shown in FIG. 17, a recess is provided on the front surface of the audio output means 5. By doing so, as shown in the figure, the flat surface 16
Even when the device is placed so that the sound output means 5 is at the bottom, it is possible to prevent the output sound from being masked on the flat surface 160 so that the listener can easily hear the sound. Become.

A voice input switch 101, a volume control switch 107, a decision switch 108, a cancel switch 109, a voice output switch belonging to the command input switch 10 described above.
The switch 110, the up switch 105, and the down switch 106 can be exchanged among them.

18, 19, and 20, the configuration including the appearance of the small-sized interpreter of the present invention is shown together. FIG.
8 is a view of the present invention as viewed from above, 602 is a view of the housing as viewed from above the side for inputting the voice of the language to be interpreted,
Reference numeral 601 is a diagram of the housing viewed from above the side surface on which the sound of the interpreted language is output.

FIG. 19 is a front view of the housing of the small-sized interpreter of the present invention as seen from all sides. 701 is a front view of the side for inputting the voice of the language to be interpreted, 702 is a side view for outputting the voice of the interpreted language, 703 is an upper side view when 702 is the front, 704 is a lower side view, 705
Is a right side view, 706 is a left side view, 707 and 708 are 70
It is the figure which looked at the inside from 5 and 706 side.

FIG. 20 is a view for explaining parts of the small interpreter of the present invention by disassembling it. 801 is a protection panel, 802
Is a screen side case, 803 is an LCD unit, 804 is an operation button, and an OK button (confirm button), BACK button (cancel button), a button for instructing audio output,
Reference numeral 5 denotes a microphone for inputting a voice language to be interpreted, reference numeral 806 denotes an operation button for scrolling a display, reference numeral 807 denotes a speaker volume dial for adjusting the output of the voice language to be interpreted, reference numeral 808 denotes a talk button for instructing the start of a translation process, and reference numeral 809 Is a memory card used as a memory for various processes, 810 is a memory card slot, 811 is a memory board holding a configuration for various processes, 812 is a speaker for outputting interpreted voice language, and 813 a small interpreter of the present invention. The power switch 814 is a speaker side case, 815 is a stored battery, and 816 is a battery lid.

However, the appearance of FIGS. 18 to 20 is merely an example, and the function of the present invention is not impaired even if the shape of the button becomes completely obscured, for example. Also,
Regarding the colors described below, a small-sized interpreter of various colors can be realized according to the user's preference, but it is clear that the function of the present invention is not affected at all.

The small interpreter of the present invention described above is intended for an interpreter that inputs a natural language by voice, converts the natural language into another natural language, and outputs the result as a voice. Since this small voice interpreter operates buttons with the thumb, the side of 701 is gray and the side of 702 is blue, pink, green, etc. in consideration of dirt.
Bright colors may be used. However, the grip portion 702 may be gray even if the grip portion 702 is different in color from other portions of the same surface in consideration of dirt.

[0044]

According to the present invention, a function of converting a language in a language input by voice through a built-in microphone into a corresponding language in another language, and outputting the conversion result as voice from a speaker is small and compact. Users with different languages can communicate with each other easily, without having to choose a time or place, without having to be aware of the language, so that they can be held in a housing that can be operated with one hand. .

Further, since the speaker for outputting the sound is installed on the surface opposite to the surface on which the microphone is installed, it is possible to prevent a howling phenomenon in which the sound output from the speaker is input to the microphone and falls into an amplification loop. It becomes possible. In addition, it is possible to proceed with the dialogue while facing each other without losing the posture or changing the way of holding the device. In addition, since the device is located at the mouth of the speaker, the utterance is interrupted, and the listener can listen only to the translated and output voice.

Further, since the power switch is installed in a portion different from the grip portion of the device, it is possible to prevent an erroneous operation in which the power switch is turned off by mistake while operating the device. Further, when the device is placed on a flat surface such that the power switch is turned down, it is possible to prevent an erroneous operation in which the power switch is turned on or off by mistake.

Further, since the center of gravity of the entire apparatus is located at the grip portion, the user can securely hold the apparatus with only one hand without perceiving a feeling of weight.

In addition, since the volume control switch is arranged so as not to be operable when the housing is placed on a flat surface, the output sound may suddenly become loud or suddenly inaudible. Obstacles can be prevented.

Further, since the microphone is installed above a portion different from the grip portion, the finger gripping the grip portion covers the microphone, thereby preventing the input voice from being masked and the voice recognition accuracy from being lowered. It becomes possible. In addition, it is possible to talk while looking at the other person's face, and it is possible to facilitate communication.

Further, since the display is installed in a portion different from the grip portion, the finger holding the grip portion covers the display to prevent the displayed characters from being masked, and the speaker can display the characters. It can be easily confirmed.

Further, since the speaker is installed in a portion different from the grip portion, a finger holding the grip portion covers the speaker to prevent the output voice from being masked, and the listener can output the voice. It becomes possible to listen easily.

Also, since the speaker is installed on the recessed portion of the housing, even when the apparatus is placed on a flat surface with the speaker positioned downward, the sound output is flat. It is possible to prevent the listener from hearing the voice easily.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a Japanese-English speech interpreter showing one embodiment of the present invention.

FIG. 2 shows an example of a data structure of a memory 6;

FIG. 3 is an external view of a Japanese-English speech interpreter showing an embodiment of the present invention, as viewed obliquely from the front.

FIG. 4 is an external view of a Japanese-English speech interpreter showing an embodiment of the present invention, as viewed obliquely from behind.

FIG. 5 is a front view and a side view of a Japanese-English speech interpreter showing an example of a breast pocket of clothes and one embodiment of the present invention.

FIG. 6 is a cross-sectional view showing the embodiment of the present invention when a Japanese-English speech interpreter is held with a right hand.

FIG. 7 is a diagram showing human fingers.

FIG. 8 is a cross-sectional view of a Japanese-English speech interpreter according to an embodiment of the present invention when the translator is stored in a breast pocket.

FIG. 9 is a side view showing the embodiment of the present invention when the Japanese-English translator is held with the right hand.

FIG. 10 is a front view and a sectional view showing the center of gravity of the Japanese-English speech interpreter showing one embodiment of the present invention.

11 shows an example of a voice recognition candidate display on the information display means 11. FIG.

FIG. 12 is an example of a language conversion result display on the information display means 11;

FIG. 13 is an example of a data structure of a language conversion table 9;

FIG. 14 is a view from the rear side of the Japanese-English speech interpreter according to the embodiment of the present invention when the right hand is used.

FIG. 15 is a cross-sectional view of a Japanese-English translator according to one embodiment of the present invention, with the front face up on a flat surface.

FIG. 16 is a sectional view of a Japanese-English speech interpreter showing one embodiment of the present invention.

FIG. 17 is a cross-sectional view of a Japanese-English translator according to an embodiment of the present invention, with a flat surface and a rear surface facing upward.

FIG. 18 is a bird's-eye view of the interpreter of the present invention.

FIG. 19 is a front view of each side of the interpreter of the present invention.

FIG. 20 is an explanatory view of parts when the housing of the interpreter of the present invention is disassembled.

[Explanation of symbols]

1 is voice input means, 2 is voice recognition means, 3 is a central processing unit, 4 is language translation means, 5 is voice output means, 6 is memory, 7
Is a voice generating means, 8 is a unit set, 9 is a language conversion table, 10 is a command input switch, 11 is information display means, 10
0 is a housing, 101 is a voice input switch, 102 is a power switch, 105 is an up move switch, 106 is a down move switch, 107
Is a volume control switch, 108 is a decision switch, 109 is a cancel switch, 110 is an audio output switch, 150 is an information input / output unit, and 151 is an operation grip unit.

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G10L 3/00 561C (72) Inventor Yasunari Obuchi 1-280 Higashi Koikebo, Kokubunji-shi, Tokyo Hitachi, Ltd. In-house (72) Inventor Yoshito Neneko 1-280 Higashi Koikekubo, Kokubunji-shi, Tokyo Inside the Hitachi, Ltd.Central Research Laboratory (72) Inventor Yukinobu Maruyama 1-280 Higashi Koikekubo, Kokubunji-shi, Tokyo In the Hitachi, Ltd.Design Research Institute (72) Inventor Hideki Uchidate 5-22-1, Kamizuhoncho, Kodaira-shi, Tokyo F-term (reference) 5C091 CB12 CB22 CB32 CD01 5D015 KK04 LL05 5D045 AA07 AB03

Claims (10)

[Claims] 1. Speech input means for inputting an uttered speech, speech recognition means for recognizing the input speech and converting it into some symbol string, and the language of the inputted speech is In a voice interpreter having language conversion means for converting into a language expression and voice output means for outputting a voice corresponding to the converted language expression, a housing is provided with a grip portion for a user to grip and operate with one hand. And an information input / output unit equipped with audio input means, audio output means, and information display means. 2. A voice interpreter according to claim 1, wherein said voice input means and said voice output means are provided on opposite surfaces of each other. 3. A housing according to claim 1, wherein said housing has a housing size and a shape that can be stored in a breast pocket of a user's clothes and can be gripped with one hand. Voice interpreter. 4. The voice interpreter according to claim 1, wherein the command input button for operation by the user is arranged so that the user can operate the command input button only with his / her finger while holding the grip portion with one hand. Machine. 5. The voice interpreter according to claim 1, wherein a power switch for operation by a user is installed in a portion different from the grip portion. 6. The grip according to claim 1, wherein:
A voice interpreter characterized by positioning the center of gravity of the entire device. 7. The voice interpreter according to claim 1, wherein the volume control switch for operation by a user is arranged so as not to be operable when the housing is placed on a flat surface. Machine. 8. The voice input device according to claim 1, wherein
A voice interpreter, which is installed above the information input / output unit. 9. The voice interpreter according to claim 1, wherein the voice output means is installed in a recessed portion of the housing. 10. An input unit for inputting a speech of a first natural language, processing means for recognizing a speech of a first natural language and converting the speech into a speech of a second natural language, A voice interpreter holding output means for outputting voice, the voice interpreter having the input unit on one side of a housing of the voice interpreter, and the input unit being provided on another side of the housing facing the one side. A speech interpreter having output means, wherein said processing means is disposed inside said housing.