JPS58156998A - Information input unit - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an information input device, and an information input device for inputting information generated by a human being, such as speech uttered in 411ζ phoneme units and handwritten characters, into a single machine.

The information input device according to the present invention can be applied whether the input information is voice or handwritten characters. However, in the following explanation, for the sake of clarity, only the case of input using voice numbers will be described. Furthermore, in the following explanation, the units of input information referred to here are Japanese monosyllables such as "a", "i", "u", "e", and "oko".

2. Description of the Related Art Conventionally, there already exist speech input devices that recognize speech uttered in units of monosyllables. As an example of this, a method has been proposed in which a monosyllable uttered in advance is registered as a standard bang, and recognition is performed by matching the monosyllable with the input monosyllable. Selection of a standard pattern is important for such a sickle-type voice input device number.

There are cases where the sound is inappropriately uttered at the time of registration, or the actual utterance may differ from the standard bang after some time has passed since registration. In order to eliminate this influence, it is effective for K to automatically learn standard bangs based on the latest input voice. If a recognition error occurs, if the user makes sure to correct it, the standard pattern can be updated based on the input voice pattern. However, in the case of a special number with a single syllable, there is a risk that the user may overlook the error, so updating in this manner may result in an incorrect speech pattern being set as the standard pattern.

SUMMARY OF THE INVENTION An object of the present invention is to provide an information input device that allows a user to correctly update a standard pattern based on a new speech pattern without having to correct a single syllable in the case of speech, for example.

The present invention includes a recognition unit that recognizes input information such as voice and handwritten characters based on a standard pattern registered in advance for each predetermined unit, and outputs a recognition result, and a word dictionary section in which words are stored; a dictionary matching section that matches the contents of the dictionary section with the above-mentioned [i* results to obtain a word recognition result; and an update instruction section for instructing pattern update.

The present invention will be described in detail below based on a specific example. Figure 1 is a block diagram of an embodiment of the present invention. In the figure, l is a microphone, 2 is an analysis section, 3 is a voice matching section as a recognition section, 4 is a standard button memory section, 5 is a dictionary matching section, 6 is a word dictionary section, and 7 is an update instruction section.

A monosyllable input through the microphone 1 is analyzed by the analysis section 2 and output as a speech pattern P. At the same time, the voice pattern P is held in the analysis section 2.

The voice pattern P is matched with the standard pattern R stored in the standard button memory section 4 in the voice matching section 3. The monosyllabic recognition result M is the certainty J
The top-ranked results are output together with a likelihood value, that is, a degree of similarity. Normally, the recognition results for monosyllables have 99% or more correct results in the top three or so, so the results to be outputted may be of this number.

The dictionary matching unit S uses the recognition result M and the word dictionary unit 6.
The word recognition result W is output by matching the contents of . The term word here does not mean the usual single @#C@, but a series of several monosyllables, and also includes phrases. When the recognition result of one word of monosyllables is input to the dictionary matching section 5, first, the word with the matching number of characters is selected and read out from the word dictionary section 6. The word in the word dictionary is compared with a monosyllable string, and the sum of the similarities in the matched monosyllables is taken as the similarity of the word. The word with the highest degree of similarity is taken as the word recognition result.

An example of this is shown in FIG. It is assumed that the larger the degree of similarity, the closer. In this case, the similarity between the words "Kanagawa" and "Kanazawa" is go+so+so+s for the former.

-2'eo the latter is 8G+SO+10+80-2201?
Therefore, the recognition result is "Kanagawa", which has a higher degree of similarity.

According to this example, it is possible to detect that the second ``na'' is mistaken for ``ma''. Based on such a result signal DB from dictionary matching 5.

The update instruction section 7 outputs an instruction signal C for updating the standard pattern in the standard button memory section 4. That is, instead of the standard pattern for "na" that is already in the standard button memory section 4, the pattern for "na" among the voice patterns P input this time and stored in the analysis section 2 is used as the standard pattern. It is held in the button memory section 4.

In addition to the above methods, several other methods can be considered for updating the standard pattern. For example, a counter is provided for each standard pattern of each monosyllable. When a recognition error of a single syllable is detected from the matching number with the dictionary, the counter of the erroneously recognized standard pattern is counted up. When the number of misrecognitions for each standard pattern exceeds a certain number of times,
When this occurs, there is a method of updating the standard C pattern with a new pattern. In addition to the method of updating the standard pattern by replacing it with a new one, there is also a method of creating a new standard pattern by taking the average of the inputted pattern and the standard pattern.

As described above, the embodiments described above are only examples selected for convenience of explanation, and the present invention is not limited to these embodiments. Various other methods can be considered for matching the input monosyllable string with the dictionary.

As mentioned at the beginning, the present invention can also be applied to handwritten character input. In this case, one character may be used as a unit instead of a single syllable. The present invention can be applied to online handwritten character recognition, etc., since it is considered that the slam matching method with standard patterns is also effective.

According to the present invention, it is possible to obtain an information input device that can update a standard pattern of a predetermined unit to a correct button in speech and character recognition.

[Brief explanation of the drawing]

FIG.

Claims (1)

[Claims] A recognition unit that registers input information such as voice and handwritten characters in predetermined units [8] based on the predetermined standard button and outputs the am result, and a recognition unit that outputs the am result as a string of the predetermined units. a word dictionary section in which words are stored; a dictionary matching section that matches the contents of the dictionary section with the pre-distributed fiber results to obtain word recognition results; An information input device comprising: an update instruction unit that determines the standard pattern and instructs updating of the standard pattern based on the determination result.