JP2001249687A - Method and system for dictation, and portable terminal device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a dictation system with a human interface capable of easily correcting a dictation result in equipment with a narrow display part. SOLUTION: A dictation result 105 in which speech inputted to a speech input part 1 is recognized by a speech recognition part 5 is divided into a morpheme string less than one row of a display part by a partial text generation part 6, and is displayed to be easily seen with the above morpheme string changed in row by a correction part 9. A partial text display part 10 can easily realize an erroneous recognition of the dictation result, the detection and correction of a wrong conversion of Kanji by easily moving the cursor to the correction point by less number of operation.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a dictation method, a dictation system, and a dictation system suitable for use in a portable terminal device requiring a display unit having a small number of characters per line, such as a cellular phone, an electronic organizer, or a car navigation system. The present invention relates to a portable terminal device used.

[0002]

2. Description of the Related Art Conventionally, dictation is performed, for example, in the form of “DIME” No.
8. As described in the September 16, 1999 issue (published by Shogakukan), it has been performed using large computers and personal computers.

Since the dictation performed by such a device is displayed on a display having a certain size, the dictation result includes many characters in one line as shown in FIG. It is easy to find misrecognition and kanji conversion in the input text in such an interface, and it has complicated input such as mouse and keyboard,
It was a technique that could easily be modified.

[0004]

As described above, dictation has been conventionally performed on a device having a relatively large display unit. However, when dictation is performed on a device having a small display unit such as a mobile phone or an electronic organizer, the dictation is performed on one line. When the dictation result is displayed on the display unit, it is difficult to find and correct erroneous recognition / kanji conversion mistake of the dictation result when the dictation result is displayed on the display unit because the number of entered characters is extremely small.

Also, for example, in a device such as a portable telephone, there is no advanced input means such as a mouse or a keyboard, and thus there is a problem that it is difficult to correct a dictation result.

SUMMARY OF THE INVENTION An object of the present invention is to provide a dictation method, a dictation system, and a portable terminal using the same that can easily detect and correct erroneous recognition of dictation results and kanji conversion errors even in a device having a display unit having a small number of characters that can be displayed in one line. It is to provide a device.

[0007]

In order to solve this problem, the present invention divides a dictation result into sub-sentences that do not extend over one line of a display unit, and changes and displays and corrects the dictation result. The object is achieved by providing means for easily moving the cursor with a small number of operations at the correction location.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The invention according to claim 1 of the present invention recognizes an input voice, creates a partial sentence from the recognition result using preset display information and grammar information, and The dictation method is characterized in that the partial sentences corrected accordingly are combined to form a sentence, and the partial sentence or the combined sentence is displayed, whereby a dictation result that is easy to see for the user using the display size information. It has the function of enabling display.

According to a second aspect of the present invention, an input means for inputting a voice, a recognition means for recognizing the input voice from the input means, and display information and grammatical information set in advance are used. A partial sentence creating unit that creates a partial sentence from the recognition result; a correcting unit that corrects the created partial sentence; a partial connecting unit that connects the partial sentences corrected by the correcting unit to form a sentence; A dictation system characterized by having a display unit for displaying a partial sentence or the connected sentence, and has an effect that the dictation result is easy to see and can be easily corrected.

According to a third aspect of the present invention, in the dictation system according to the second aspect, the display means separates an area for displaying a partial sentence from an area for displaying a sentence connecting the partial sentences. And has the effect of facilitating confirmation and correction of the dictation result.

[0011] According to a fourth aspect of the present invention, in the dictation system according to the second aspect, the display information is the number of characters that can be displayed on one line. This has the effect of displaying a suitable dictation result.

According to a fifth aspect of the present invention, in the dictation system according to any one of the second to fourth aspects, the partial sentence has a basic unit of a morpheme which is a minimum unit having a meaning. This has the effect of making it easy to understand the meaning of the partial sentence and making the correction easy.

According to a sixth aspect of the present invention, in the dictation system according to the second aspect, the grammatical information is characterized in that the beginning and end of the sub-sentence are determined by a classification result such as a morpheme, such as a part of speech. And has the effect that the meaning of the sub-sentence can be immediately understood.

According to a seventh aspect of the present invention, there is provided an example sentence storing means for storing a plurality of example sentences, and a partial sentence converting means for converting the example sentences into partial sentences using preset display information and grammatical information. Correction input means for moving a cursor to a correction position of the partial sentence, input means for inputting voice, recognition means for recognizing input voice from the input means, and a part for creating a partial sentence from the recognition result Sentence creating means, correcting means for correcting a partial sentence located at the correction position by the created partial sentence, partial connecting means for connecting the partial sentence corrected by the correcting means to form a sentence, A dictation system characterized by having a display means for displaying a sentence or the connected sentence, and has an operation of easily editing a stored example sentence.

According to an eighth aspect of the present invention, in a portable terminal device using the dictation system according to any one of the second to seventh aspects, a communication means for transmitting and receiving connected sentences is provided. The portable terminal device has an operation of providing an easy dictation operation in the portable terminal.

According to a ninth aspect of the present invention, in the portable terminal device according to the seventh aspect, the example sentence storage means stores sent / received sentences as example sentences. Has an easy sentence creation effect.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.

(Embodiment 1) FIG. 1 is a configuration diagram of an example of a portable terminal device using a dictation system according to Embodiment 1 of the present invention.

In FIG. 1, 1 is a voice input unit for inputting voice and converting it into an electric signal, 2 is a feature extracting unit for converting the input voice into acoustic parameters, and 5 is a sound converted by the feature extracting unit 2. The voice recognition unit 16 recognizes using the parameters and the acoustic model 3 and the language model 4 obtained in advance. A possible example sentence storage unit 6 is a partial sentence creation unit that creates a partial sentence from the recognition result or the selection result of the example sentence storage unit 16 using display information 7 and grammar information 8 set in advance. A correction unit that corrects a partial sentence based on information input by the information input unit 11 other than the voice, and a partial sentence connection unit that connects the partial sentences corrected by the correction unit 9 to form a sentence, and 12 Is part Or display unit for displaying the linked text, 15 is a communication unit that transmits the connection statement wirelessly or wired, and the like.

The operation of the portable terminal device using the dictation system configured as described above will be described below.

The voice input to the voice input unit 1 is converted by the voice input unit 1 into an input voice signal 101 as an electric signal, and is converted by the feature extraction unit 2 into appropriate voice parameters 104 such as LPC cepstrum coefficients. . The voice recognition unit 5
The acoustic parameters 104 obtained by the feature extracting unit 2 and the acoustic model data 10
2. Recognize input speech using language model data 103 from language model 4. The partial sentence creation unit 6 determines the number of characters of the partial sentence using the number of characters that can be displayed in one line from the display information 7 on the dictation result 105 that is the recognition result of the speech recognition unit 5 or the sentence extracted from the example sentence storage unit 16. Using display information data 106 to be controlled and grammatical information data 107 for performing grammatical restrictions such as not ending a partial sentence with a noun when following a noun or a particle from the grammatical information 8, a plurality of morphological units are used as a minimum unit. Partial sentence signal BS10
Converted to 8. The partial sentence signal BS108 is sent to the correction unit 9, and the resulting partial sentence signal 109 in the corrected state is output.
Is displayed by changing the line for each partial sentence signal in the partial sentence display unit 10.

The example sentence stored in the example sentence storage unit 16 is a speech recognition result, a key input, and a sentence received from the communication unit, which can be used.

The display information such as the number of characters of the partial sentence required in the display information 7 is not limited to the one fixed by hardware, but can be changed by software or external information.

Further, the partial sentence signal BS108 is displayed on the partial sentence display section 10 at the beginning of each line, for example, as shown in FIG. 2, and a mark (for example, /) 24 for separating each morpheme 23 is entered. Also, when a partial sentence is displayed on the partial sentence display section 10, the cursor 25 is automatically set at the head of the uppermost partial sentence which is the foremost of the dictation result.

The correction unit 9 then uses the cursor to point out a part that corrects erroneous recognition of dictation results and kanji conversion errors displayed in units of partial sentences, and, for example, uses keys, buttons, and jog dials other than voices. The correction information signal 11 from the information input unit 11 or the speech recognition unit 5
Enter 0.

Here, a specific correction method will be described. In order to distinguish between a case where correction is performed in units of partial sentences displayed in each line and a case where correction is performed in units of the first morpheme of a partial sentence displayed in each line, for example, when indicating a partial sentence displayed in each line, use a partial When the cursor is placed on the numeral in front of the sentence and the morpheme at the head of the partial sentence is to be corrected, the cursor is placed on the mark separating each morpheme and then the correction operation is performed.

The partial sentence display section 10 displays the partial sentence by changing the line. Therefore, when it is desired to correct erroneous recognition of a dictation result or kanji conversion error in partial sentence units, move the cursor downward as it is and correct it. Move the cursor to the sub-sentence you want to use or specify the line number, move the cursor to the sub-sentence, and in the case of misrecognition, re-utter the sub-sentence or type in characters, etc. In the case of a kanji conversion error, bring the cursor in front of the correction section, display conversion candidates, and select a correct kanji.

When it is desired to modify a partial morpheme in a partial sentence, the cursor is moved to the position before the partial sentence including the morpheme to be modified, and the cursor is moved to the right in morpheme units. Can be moved to a location. Also, since there are not so many morphemes in a partial sentence, it is easy to see at what line the morpheme to be corrected is located from the beginning of the partial sentence. So, for example, if there is a number button like a mobile phone, move the cursor to the morpheme that you want to easily modify by entering the line number and the number of the morpheme from the beginning of the partial sentence, Corrections can be made. Note that the two numbers indicating this morpheme can also be obtained by voice recognition.

When the correction in the correction section 9 is completed, a fixing operation is performed. In this, a dictation result 112 corrected by concatenation of partial sentences is created in a partial sentence linking unit 14 for linking the dictation results 111 divided into partial sentences, and is sent to the display unit 13 for confirmation. Confirmation display unit 13
Displays the newly determined dictation result following the previously entered text.

The determined sentence 113 is transmitted by the communication unit 15 using a wired or wireless communication.

It should be noted that the partial sentence display section 10 and the confirmation display section 1
For example, as shown in FIG. 3, the display unit of the dictation system may be displayed in two parts, as shown in FIG. 3, or the dictation result may be corrected as shown in FIG. Only the part is displayed, and after performing the correction operation after correction, only the display part for determination is displayed on the display part of the dictation system.

Here, the confirmation display section is, for example, a display screen of an application of mail software or word processing software, and can transmit a dictation result to a device having another display size.

[0033]

As described above, according to the present invention, the dictation result can be easily distinguished and the cursor can be easily moved to the correction position for a device having a display mode in which the number of characters per line is small. Therefore, the user can easily correct his / her own utterance result, so that even a device having only a narrow display unit can realize a human interface that can easily create sentences by dictation.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a dictation system according to Embodiment 1 of the present invention.

FIG. 2 is a display screen diagram showing a display example of a partial sentence display unit in the embodiment.

FIG. 3 is a display screen diagram showing a display example in which a partial sentence display unit and a display unit for confirmation are simultaneously displayed in the embodiment.

FIG. 4 is an exemplary display screen showing a display example in which a partial sentence display unit and a confirmation display unit are displayed separately according to the embodiment;

FIG. 5 is a diagram showing a display example of the related art

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Speech input part 2 Feature extraction part 3 Language model 4 Acoustic model 5 Speech recognition part 6 Partial sentence creation part 7 Display information 8 Grammar information 9 Correction part 10 Partial sentence display part 11 Information input part other than voice 12 Display part 13 Confirmation Display unit 14 partial sentence connection unit 15 communication unit 16 example sentence storage unit 101 input speech signal 102 acoustic model data 103 language model data 104 acoustic parameters 105 dictation result 106 display information data 107 grammar information data 108 partial sentence signal 109 part in a corrected state Sentence signal 110 Correction information signal 111, 112 Dictation result 113 Determined sentence

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G10L 15/00 G10L 3/00 551B 561C (72) Inventor Akira Ishida 3-chome, Higashi-Mita, Tama-ku, Kawasaki-shi, Kanagawa Prefecture. No.10 No.1 Matsushita Giken Co., Ltd. F-term (reference) 5B009 KB04 RB01 5D015 AA05 KK02 5E501 AA04 AA11 AA23 AB03 AC34 EB15 EB17 EB19 EB20 FA02 FB22 9A001 HH15 HH17 JZ77

Claims (9)

[Claims] An input speech is recognized, a partial sentence is created from the recognition result using preset display information and grammatical information, and a corrected partial sentence is connected as necessary to form a sentence. A dictation method comprising displaying the partial sentence or the connected sentence. 2. An input unit for inputting a voice, a recognition unit for recognizing an input voice from the input unit, and a part for creating a partial sentence from the recognition result using preset display information and grammatical information. Sentence creating means, correcting means for correcting the created partial sentence, partial connecting means for connecting the partial sentences corrected by the correcting means to form a sentence, and displaying the partial sentence or the connected sentence A dictation system comprising: 3. The dictation system according to claim 2, wherein the display means separates an area for displaying the partial sentence from an area for displaying a sentence connecting the partial sentences. 4. The dictation system according to claim 2, wherein the display information is the number of characters that can be displayed on one line. 5. The sub-sentence according to claim 2, wherein a basic unit is a morpheme which is a minimum unit having a meaning.
The dictation system according to any one of the above. 6. The dictation system according to claim 2, wherein the grammatical information determines the beginning and end of the sub-sentence based on a classification result of a morpheme such as a part of speech. 7. An example sentence storage means for storing a plurality of example sentences,
Using a display information and grammar information set in advance of the example sentence, a partial sentence conversion unit that converts the sentence into a partial sentence, a correction input unit that moves a cursor to a correction position of the partial sentence, and an input unit that inputs a voice, Recognizing means for recognizing an input voice from the input means, partial sentence creating means for creating a partial sentence from the recognition result, and correcting means for correcting a partial sentence located at the correction position by the created partial sentence; A dictation system comprising: a partial connecting unit that connects the partial sentences corrected by the correcting unit to form a sentence; and a display unit that displays the partial sentence or the connected sentence. 8. A portable terminal device using the dictation system according to claim 2;
A mobile terminal device comprising communication means for transmitting and receiving a connected sentence. 9. The portable terminal device according to claim 7, wherein the example sentence storage unit stores the sent / received sentence as an example sentence.