JPH10171492A - Electronic dictionary device and operation method therefor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce button operations and the number of buttons and to improve operability by adopting duration of pushing a speech recognition button and the number of times as parameters. SOLUTION: A speech is inputted while pushing a speech recognition button 13 for one second or longer. If a speech-inputted word is the same as a word displayed on a liquid crystal display 10, consulting a dictionary by the speech recognition is successful. If a speech-inputted word differs from a word displayed on the liquid crystal display 10, the next candidate for the speech recognition is displayed by pushing the speech recognition button 13 once within 0.5 seconds. If a candidate is outputted up to a third one but the speech-inputted word differs from the word displayed on the liquid crystal display 10, the same word is speech-inputted again by pushing the speech recognition button 13 for one second or longer. When dictionary consultation is interrupted, the speech recognition button 13 is pushed two times within 0.5 seconds during the state when the device is waiting for a next instruction.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an electronic dictionary device having a voice recognition function. In particular, the present invention relates to an operation method for effectively using a voice recognition function for an electronic dictionary device.

[0002]

2. Description of the Related Art In recent years, electronic technology has been developed, and a 70,000-word Japanese-English dictionary and the like have appeared in a notebook-sized electronic dictionary. Some of them contain many explanations and example sentences, which can be used as detailed dictionaries.

The method of operating this electronic dictionary is usually a push button method, and is performed by using alphabet buttons or menu-related buttons. When using an electronic dictionary on a desktop, button operations are not difficult compared to when carrying, but when carrying, one hand is often blocked and both hands cannot be used, so many button operations Not suitable for doing.

In the case of an electronic dictionary device, a dictionary lookup operation is performed using an alphabet button. However, this is not suitable for real-time operation. For example, in the case of a Japanese-English dictionary, an operation method in which an alphabet button or a kana button is pressed, a kana input is performed, and then a Japanese-English conversion is performed takes time.

[0005] Elements that require a long time to perform a dictionary lookup operation are small in alphabetical buttons and large in the difficulty of operation. As with a typewriter, a desktop-sized computer or word processor keyboard can perform dictionary lookup in near real time by getting used to operation with both hands. In this regard, computer-friendly dictionary software has begun to be released.

FIG. 4 is a diagram showing the appearance of a conventional electronic dictionary device. The electronic dictionary device shown in FIG.
It comprises an FF button 40, a menu-related button 41, an alphabet button 42, a conversion button 43, an arrow button 44, a liquid crystal display 48, and a speaker 49. The menu-related buttons include an English-Japanese button 45, a Japanese-English button 46, and an utterance button 47.

Next, a method of operating the conventional electronic dictionary device shown in FIG. 4 will be described. First, the power ON / OFF button 40
To make it operable. Press the Japanese-English button 46 in the menu button. Then, a Japanese-English dictionary mode is entered, and here, the kana of the word to be looked up in the dictionary is input while pressing the alphabet button 42 to perform the Roman alphabet kana conversion.
If an input error is made, the user presses the left arrow button 44 to delete and input again. FIG. 5 shows the result of inputting “Kana” in the liquid crystal display of FIG. Thereafter, the conversion button 43 is pressed to perform Japanese-English conversion.
The dictionary lookup result is displayed on the liquid crystal display 48. If the description of the dictionary lookup result does not fit within one screen, the user performs a scroll operation with the up and down arrow buttons 44 to operate so as to see a necessary part.

When the user wants to utter a message in the screen, he presses a utterance button 47 in the menu button. Then, words and sentences in which the utterance function is registered are uttered from the speaker 49 in the description of the dictionary lookup result, and the pronunciation method is known.

Such an operation method is a conventional technique. In particular, in a situation where a Japanese person immediately searches for an English word he or she wants to know during an actual conversation, the user must perform dictionary lookup while pressing several buttons. It requires a fairly quick button operation.

[0010] The same operation method is used when a dictionary is searched by pressing the English-Japanese conversion button. The difference is that the alphabet of the word to be looked up in the dictionary can be input as it is.

[0011]

As described above, in the operation method of the conventional electronic dictionary device, in both the Japanese-English dictionary mode and the English-Japanese dictionary mode, the input operation of inputting with the alphabet button cannot be avoided.
There is a problem that it takes a long time to input a dictionary for use in real time.

In addition, there are too many buttons in the apparatus, and when used quickly, there is a problem that the user mistakenly presses the button, temporarily forgets where the button is, and confuses the user.

[0013] Furthermore, there is a problem that one hand cannot be used for practical use when carrying.

It is an object of the present invention to solve the above-mentioned problems, and to provide an excellent method of operating an electronic dictionary device in which button operations and the number of buttons are reduced and operability is improved.

[0015]

In order to achieve the above object, an electronic dictionary device operating method according to the present invention comprises a liquid crystal display, a power ON / OFF button, a dictionary type selection button, and a voice input button. Recognition button, voice button,
In an electronic dictionary device including a microphone and a speaker, the number of voice recognition buttons is one, and voice recognition, result output, and voice recognition function interruption are performed using the time and the number of times the voice recognition button is pressed as a parameter. And

Further, the method of operating the electronic dictionary device of the present invention comprises:
While the user presses the voice recognition button for a certain period of time or longer, the user utters a word to be searched in the dictionary, voice recognition is performed, the first candidate of the recognition result is displayed on the liquid crystal display, and voice recognition is performed. When the voice recognition button is pressed for a time equal to or less than half of a certain time or more and for a certain time or less, a second candidate is output, and the third candidate and the fourth candidate are similarly set to the device. It is characterized in that up to candidates are output.

Further, the method of operating the electronic dictionary device of the present invention comprises:
Pressing the button for performing voice recognition is half or less than a certain period of time or longer, and pressing the voice recognition button for a certain period of time or less causes candidates to appear one after another. If a certain number of candidates are output and the same word as the user uttered is not displayed on the liquid crystal display, then the user presses the voice recognition button for a certain period of time or more, and the user searches the dictionary for the same word. Is uttered, and when speech recognition is executed, there is a method in which, in the result output, all candidates in the previous speech recognition are not included in the current candidates.

Further, in the operation method of the electronic dictionary device according to the present invention, while the user presses the voice recognition button for a certain period of time or more, the user utters a word to be looked up in the dictionary, voice recognition is executed, and the first candidate of the recognition result is obtained. Is displayed on the liquid crystal display, and when a certain number of candidates set for the device are displayed, the time required to press the voice recognition button for performing voice recognition is longer than a certain time. When the voice recognition button is pressed twice or more for less than half and for a certain time or less, the voice recognition function of the voice recognition electronic dictionary is interrupted.

[0019]

FIG. 1 shows the external appearance of an electronic dictionary device according to an embodiment of the present invention. Further, the configuration of the electronic dictionary device shown in FIG. 1 will be described below.

An electronic dictionary device according to the present invention shown in FIG. 1 includes a liquid crystal display 10, a power ON / OFF button 11, a menu-related button 12, and a voice recognition button 1.
3, screen scroll buttons 14, 15 and microphone 16
And the speaker 17. Menu-related buttons include English-Japanese button 18, Japanese-English button 19, and voice button 20
It consists of.

A power ON / OFF button 11 is used to turn on and off the power of the apparatus. A menu-related button 12 is an English-Japanese button 18 for switching the dictionary type.
A Japanese-English button 19 and an utterance button 20 for uttering words and sentences for explanation of the dictionary lookup result. The screen scroll buttons 14 and 15 are used when the description of the dictionary lookup result is long.
By operating these two screen scroll buttons, the explanation is moved up and down to view the whole sentence. The voice recognition button 13 is used to perform dictionary lookup by voice. In the case of voice recognition, since the voice input of the user cannot be correctly recognized 100% reliably, the second candidate, the third candidate, or the like is output. This button allows the user to forcibly escape from the voice recognition function on the way.

Among the above buttons, the operation method related to the power ON / OFF button 11, the menu related buttons 12, and the screen scroll buttons 14, 15 is the same as the conventional example.

Hereinafter, an example of a method of operating the electronic dictionary device using the voice recognition button 13 of the present invention will be specifically described.

The Japanese-English button 19 is pressed to enter the Japanese-English dictionary mode. Thereafter, a Japanese word to be looked up in the dictionary is input by voice. A method of inputting a voice while pressing the voice recognition button 13 for 1 second or more is adopted. Thereafter, as a result of the speech recognition, a Japanese display of the word and an English description of the word are displayed on the liquid crystal display 10. If the word input by voice is the same as the word displayed on the liquid crystal display 10, the dictionary lookup by voice recognition has been successful, and immediately after reading the explanation, the The user may take the following countermeasures, for example, by performing a dictionary lookup of the word "".

However, if the word input by voice is different from the word displayed on the liquid crystal display 10,
Dictionary lookup by voice recognition has not been successful,
The next candidate of the voice recognition result is displayed by pressing the voice recognition button once within 0.5 seconds. The Japanese display of the second candidate of the speech recognition result and the English description of the word are displayed on the liquid crystal screen. Here, if the word input by voice is the same as the word displayed on the liquid crystal screen, the user may respond in the same manner as described above.

If the word input by voice is different from the word displayed on the liquid crystal display 10 again, the next candidate for the voice recognition result is displayed in the same manner as described above. Even with the latest technology of speech recognition, it is better to set the apparatus so as to output up to the third candidate, and in the present embodiment, it is also possible to output up to the third candidate. Any number can be set.

Further, even when the third candidate is output, there is a case where the word input by voice differs from the word displayed on the liquid crystal screen. When the input voice uttered by the user has an ambiguous pronunciation, when the environment of the place contains noise, etc., and adversely affects speech recognition, or when a word that has many similar words is uttered, Such things are easy to happen. In consideration of such a case, if the word input by voice and the word displayed on the liquid crystal display 10 are different even when the third candidate is output, the voice recognition button 13 is pressed again for one second or more to repeat the same word. Speech is input, and the speech recognition result is displayed without including candidates for the previous speech recognition result.

Next, a description will be given of a method of interrupting a dictionary lookup by voice recognition of a word. When the apparatus is in a state of waiting for the next instruction, the voice recognition button 13 is pressed twice within 0.5 seconds. In this case, the apparatus recognizes that the dictionary lookup of the word being performed has been interrupted, and the dictionary lookup of another word has been performed.

FIG. 2 shows the timing of dictionary lookup using speech recognition according to the present invention. It is a figure for showing a timing relation between operation of a voice recognition button, and a user's voice input. At 20 in FIG. 2, while the voice recognition button is pressed for one second, the voice of "airport" is input. Similarly, 21 and 2
In 2, 25, "international line", "non-smoking seat", and "takeoff" are input by voice, the voice recognition function is activated, and the dictionary is searched. At 23, the third candidate is output, and at 24, "non-smoking seat"
It can be seen that the speech recognition function has been interrupted for the dictionary lookup of the word.

The speech recognition method used in this speech recognition electronic dictionary is basically based on HMM (Hidden Mar
cov Model) and the One-Pass-DP method.

Briefly described with reference to FIG. 3, an input voice uttered while the user presses the voice recognition button 34 for dictionary lookup enters the microphone 30 in the apparatus, becomes an analog electric signal by the microphone, and then cuts off. Frequency is 5kHz
A / D through a low-pass filter (LPF) 31 of z
The data is A / D converted to 16 kHz and 16 bits by the converter 32, converted into a digital electric signal, and stored in the memory 33 in the apparatus. In parallel, the data stored in the memory 33 is calculated by using the CPU 35 as well.
The data is converted into a feature amount called an LPC mel-cepstrum and stored in the memory 33. Furthermore, in parallel with it, L
PC mel cepstrum data is HMM & One-Pa
In the ss-DP method, further calculation processing is performed, and recognition results are output from the first candidate to the designated candidate (third candidate in the present embodiment).

[0032]

As described above, according to the present invention, it is not necessary to operate the alphabet buttons to look up the dictionary as in the conventional example, and the dictionary can be looked up by a simple operation. Also,
Since the dictionary can be looked up in real time, it is very effective to use it for actual conversation. Furthermore, since a dictionary lookup based on the voice recognition function in consideration of the actual technical level of voice recognition has been devised, an electronic dictionary with excellent operability that can be easily viewed as the second and third candidates can be provided. it can.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating an appearance of a voice recognition electronic dictionary according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating timings of a voice recognition button and a voice input of the electronic dictionary device according to the embodiment of the present invention.

FIG. 3 is a block diagram of hardware related to a voice recognition function in the embodiment of the present invention.

FIG. 4 is an external view showing a configuration of a conventional electronic dictionary device.

[Explanation of symbols]

 10 LCD Display 11 Power ON / OFF Button 12 Menu Related Buttons 13 Voice Recognition Button 14 Screen Up Scroll Button 15 Screen Down Scroll Button 16 Microphone 17 Speaker 18 English-Japanese Button 19 Japanese-English Button 20 Voice Button

Continued on the front page (51) Int.Cl. ⁶ Identification code FI G06F 3/16 340 G06F 3/16 340R 15/02 325 15/02 325B 17/30 15/40 370J // G06F 17/28 15/403 310Z 15/38 C

Claims (4)

[Claims] 1. A liquid crystal display and power ON / OFF
, A dictionary type selection button for performing dictionary lookup, a voice recognition button, a voice utterance button, an electronic dictionary device having a microphone and a speaker, wherein the voice recognition button is one, An electronic dictionary device for performing voice recognition, outputting a result, and interrupting a voice recognition function by using a time and a number of times of pressing a voice recognition button as parameters. 2. An operation method for performing voice recognition and outputting a result by pressing a voice recognition button of the electronic dictionary device according to claim 1, wherein the voice recognition button is pressed for a predetermined time or more.
The user utters the word to be searched in the dictionary, voice recognition is performed, the first candidate of the recognition result is displayed on the liquid crystal display, and the voice recognition button is pressed for performing the voice recognition. When the voice recognition button is pressed for less than half and for a certain time or less, a second candidate is output, and similarly, a third candidate, a fourth candidate, and a candidate set in the device are output. How to operate the electronic dictionary device. 3. A method of performing voice recognition and outputting a result by pressing a voice recognition button of the electronic dictionary device according to claim 1, wherein the time is longer than a certain time when the voice recognition button for performing the voice recognition is pressed. By pressing the voice recognition button for less than half and for a certain time or less, candidates are sequentially output, but a certain number of candidates set in the device are output, and the same as the word uttered by the user is output. If the word is not displayed on the liquid crystal display, then the user presses the voice recognition button for a certain period of time or more, and the user utters the same word to be looked up in the dictionary. And a method of operating the electronic dictionary device, wherein all candidates in the previous speech recognition are not included in the current candidates. 4. A method for interrupting a voice recognition function by pressing a voice recognition button of an electronic dictionary device according to claim 1, wherein the voice recognition button is pressed for a predetermined time or more.
When the user utters a word to be looked up in the dictionary, voice recognition is performed, and the first candidate of the recognition result is displayed on the liquid crystal display, and when a certain number of candidates set in the apparatus are displayed. In the electronic dictionary device, when the voice recognition button is pressed twice or more for half or less of a certain time or more and for a certain time or less for pressing the voice recognition button for performing the voice recognition, A method for operating an electronic dictionary device, wherein the function is interrupted.