JP3621624B2 - Foreign language learning apparatus, foreign language learning method and medium - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a foreign language learning apparatus and a foreign language learning method for performing foreign language learning using a speech recognition system, and a computer-readable medium recording a program for causing a computer to execute such a foreign language learning method. About.
[0002]
[Prior art]
In recent years, many attempts have been made to use speech recognition systems for learning foreign languages. In other words, in such a foreign language learning device, the learner reads out one or more foreign language sentences and inputs his / her pronunciation into the personal computer using the voice input function of the personal computer (computer). To do. A method in which the speech recognition system corresponding to the foreign language installed in the personal computer determines how correctly the sentence read by the learner is recognized and displays the result as feedback to the learner. It is.
[0003]
[Problems to be solved by the invention]
However, such a learning method can only determine the quality of pronunciation of the spoken foreign language as a whole.
[0004]
Actually, it is necessary to improve the pronunciation ability of foreign language sounds that do not exist in Japanese, in order to improve the effect by practicing pronunciation closer to foreign language sounds.
[0005]
In order to listen to the pronunciation of a foreign language and make the corresponding pronunciation exactly yourself, for example, the difference between the sound system of the Japanese language and the sound system of the foreign language you learn is the native language of Japanese. It is a big obstacle when people learn foreign languages.
[0006]
Based on recent research results on foreign language learning, it has been found that the difficulty of learning foreign language sounds depends on the learner's mother tongue. In particular, listening and pronunciation of sounds not present in the mother tongue is particularly difficult for learners. A major cause of this difficulty is that the human speech-language information processing mechanism that normally humans mistakenly recognize sounds that are not in their native language as sounds close to that in their native language. The characteristics are listed.
[0007]
For example, it is a typical example that a Japanese hears “red” and “led” in English as “red” in Japanese, and pronounces both of them as “red”. .
[0008]
In other words, when learning a foreign language sound, the learner's listening and utterance are strongly bound by the sound system of the native language. In particular, there is a problem that it is difficult to improve speech ability.
[0009]
The present invention has been made to solve the above-mentioned problems, and its purpose is to train pronunciation of sounds peculiar to foreign languages that do not exist in the native language as a natural extension of the native sound system. A foreign language learning apparatus and a foreign language learning method capable of efficiently performing the above.
[0010]
[Means for Solving the Problems]
The foreign language learning device according to claim 1, further comprising storage means for holding an extended phonetic table to be pronounced by a learner and corresponding model phoneme information, wherein the extended phonetic table is at least in a Japanese syllable table. It is a sound table to which either a consonant or a vowel peculiar to the foreign language to be studied is added, and an output means for presenting at least a part of the extended sound table to the learner in advance, and voice information that the learner has pronounced The phoneme separation means for separating the phoneme speech information for each phoneme, and the evaluation means for evaluating the degree of coincidence with the model phoneme information for each phoneme speech information, and the display device displays the determination result of the evaluation means. It is displayed at least for each phoneme and syllable.
[0011]
The foreign language learning device according to claim 2 is the configuration of the foreign language learning device according to claim 1, wherein the extended sound table is obtained by adding [r] sound and [l] sound to the vowel array side. .
[0012]
The foreign language learning method according to claim 3, wherein the learner uses at least a part of an extended phonetic table in which at least one of a consonant and a vowel peculiar to the foreign language to be learned is added to the Japanese alphabet. Presenting as syllables to be pronounced, receiving speech information generated by the learner, separating into phoneme speech information for each phoneme, and evaluating a degree of coincidence with model phoneme information for each phoneme speech information And a step of displaying the evaluation result at least for each phoneme and for each syllable.
[0013]
The foreign language learning method according to claim 4 is the configuration of the foreign language learning method according to claim 3, wherein the extended phonetic table is obtained by adding [r] sound and [l] sound to the vowel array side. .
[0014]
6. A computer-readable medium recording a program for causing a computer to execute the foreign language learning method according to claim 5, wherein the foreign language learning method is applied to at least a foreign language to be learned in a Japanese syllabary table. A step of presenting at least a part of an extended syllab to which either a specific consonant or a vowel is added as a syllable to be pronounced by the learner, and a phonetic speech information for each phoneme in response to the speech information pronounced by the learner And the step of evaluating the degree of coincidence with the model phoneme information for each phoneme speech information, and the step of displaying the evaluation result at least for each phoneme and for each syllable.
[0015]
According to a sixth aspect of the present invention, in the medium configuration of the fifth aspect, the extended sound table is obtained by adding [r] sound and [l] sound to the vowel array side.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a schematic block diagram for explaining the configuration of a foreign language learning apparatus 100 according to the present invention.
[0017]
In the following description, a case where a learner whose native language is Japanese learns English as a foreign language will be described as an example. However, as will be apparent from the following description, the present invention is in such a case. The present invention is not limited to the above, and is generally applicable when a learner whose native language is Japanese learns a foreign language other than English.
[0018]
Referring to FIG. 1, foreign language learning apparatus 100 has a microphone 102 for acquiring a voice uttered by learner 2, and a voice corresponding to a syllable pronounced by learner 2 in response to an output from microphone 102. For each phoneme included in this syllable, the information is processed, and the learner 2 given by the microcomputer 110 pronounces the learner's pronunciation evaluation score based on the expected pronunciation. A display device (display) 120 is provided for displaying a power table and syllables, and displaying evaluation scores for learners' pronunciations obtained for each phoneme and each syllable.
[0019]
In addition, as a method of presenting the syllable to be pronounced (hereinafter referred to as “model syllable”) to the learner 2, as described above, it may be presented as character information via the display device 120. You may show to the learner 2 as an audio | voice from the speaker 104. FIG. Further, in pronunciation practice for each word, which will be described later, it is possible to output a pronunciation as a model from the speaker 104 as speech.
[0020]
The microcomputer 110 receives an audio signal from the microphone 102 and also provides an audio input / output unit 112 which is an interface for giving an audio signal to the speaker 104, and a signal from the audio input / output unit 112 as will be described later. Based on the speech recognition unit 114 that separates and analyzes speech information corresponding to the syllable given to the microphone 102 (hereinafter referred to as “syllable speech information”) into phonological information included therein, and temporarily converts the syllable speech information And the data storage unit 118 for storing the model syllable, the model phonological information corresponding to the model syllable, and information on the phoneme boundary, the analysis result of the speech recognition unit 114, and the data storage unit 118. This model is based on the information about the model syllable given to encourage the learner 2 to pronounce. A phoneme and per each syllable contained in syllables, and a processing unit 116 for determining an evaluation point pronunciation learner 2 for the phoneme information for the model syllables (Model phoneme information).
[0021]
[Japanese sound system]
FIG. 2 is a sound table for explaining the Japanese sound system.
[0022]
Kana in Japanese is a phonetic symbol, and has excellent characteristics as a system for expressing sounds. For this reason, in general, a Japanese speaker can easily pronounce even a meaningless syllable (arrangement of vowels and consonants).
[0023]
On the other hand, for example, if Americans do not write their pronunciation in spelling but represent them with phonetic symbols, it would be a very difficult task.
[0024]
Therefore, for Japanese speakers, it is possible to practice foreign languages by expanding the Japanese phonetic table (50-syllabary table) when using the fact that kana is a phonetic symbol and making pronunciation based on characters. It is possible to do.
[0025]
Here, as shown in FIG. 2, the Japanese syllabary table has five vowels “a”, “i”, “u”, “e”, “o” vertically, and “none” (No) ”,“ k ”,“ s ”,“ t ”,“ n ”,“ h ”,“ m ”,“ y ”,“ r ”,“ w ”, 10 consonants. ing.
[0026]
[Expansion of the Japanese syllabary table for foreign language pronunciation practice]
In the present invention, the data storage unit 118 of the foreign language learning apparatus described with reference to FIG. 1 stores expanded phonetic table data and model pronunciation (model phoneme information) for the phonetic table as described below. ing.
[0027]
FIG. 3 shows, for example, an expanded phonograph for practicing English consonants. In English / r /, / l / consonant concatenation with sounds in it, for example / kr / at the beginning of “cry”, / bl / at the beginning of “black” etc. are particularly difficult for Japanese learners It is an easy learning object.
[0028]
The reason is that, in Japanese, the so-called CV format (C: consonant, V: vowel) where the vowel is always connected after the consonant is taken, the CC format in which two consonants are continuous is Because of its phonological structure, it is difficult for Japanese to pronounce. In addition, it is thought that two factors that the / r /, / l / sound itself is difficult for Japanese people overlap.
[0029]
Therefore, in the phonogram shown in FIG. 3, for the practice of pronunciation of consonant concatenation including / r / and / l /, the fact that / r / and / l / are both strong vowels is applied. In the note table shown in Fig. 3, these pronunciations are also regarded as vowels.
[0030]
By adopting such a configuration, pronunciation learning of consonant concatenation including / r / and / l / is efficiently performed by performing pronunciation practice as Cr (consonant + / r /) and Cl (consonant + / l /). Is possible.
[0031]
That is, referring to FIG. 3, not only the combination of Japanese vowels a, i, u, e, o, etc. and Japanese consonants k, etc., but also Japanese vowels a, i, u, e, By practicing the combination of o and the vowel θ of English, it is possible to first practice pronunciation of these consonants.
[0032]
Furthermore, it is possible to practice consonant concatenation by practicing pronunciation of “kr”, “kl”, etc., which are combinations of Japanese consonant k and consonant R or L.
[0033]
At this time, in the foreign language learning apparatus shown in FIG. 1, the pronunciation made by the learner is collected by the microphone 102 and analyzed by the voice recognition unit 114, thereby storing the model stored in the data storage unit 118. By comparing with the sound, the evaluation result for the pronunciation of these consonants can be calculated by the arithmetic processing unit 116 and displayed on the display device 120.
[0034]
Furthermore, in order to make the pronunciation practice easier, the pronunciation practice can be performed as a phrase obtained by adding “m”, “n”, “ng”, etc. as a consonant after the consonant + vowel. For example, pronunciation as a word such as “wam”, “wan”, “wang” can be performed.
[0035]
[How to practice pronunciation]
FIG. 4 is a flowchart for explaining a foreign language learning flow realized by foreign language learning apparatus 100 shown in FIG.
[0036]
As will be apparent from FIG. 4, in the foreign language learning realized by the foreign language learning device 100, intensive pronunciation practice for phonemes, which is not good for Japanese people, is concentrated based on the expanded phonetic table described above. It is a configuration that can be performed.
[0037]
Referring to FIG. 4, when foreign language learning is started (step S100), first, display device 120 presents an expanded sound table to be pronounced to learner 2, and should be pronounced in this sound table. A syllable is presented (step S102). At this time, the syllables to be pronounced may be designated in the foreign language learning apparatus 100 in the order according to the phonetic table or may be designated randomly. Furthermore, according to a learner's designation | designated, it is good also as a structure which designates repeatedly intensively about the syllable or the syllable group which the learner determined not to be good based on the evaluation result demonstrated later.
[0038]
Subsequently, when the learner 2 pronounces the model syllable, voice information (sentence voice information) corresponding to the model syllable is acquired via the microphone 102 and the voice input / output unit 112 (step S104).
[0039]
Next, the speech recognition unit 114 recognizes the text speech information as speech information corresponding to each phoneme based on the signal output from the speech input / output unit 112 (step S106).
[0040]
Subsequently, the arithmetic processing unit 116 obtains an evaluation score by referring to the model phonological information for the model syllable stored in the data storage unit 118 for each syllable or for each phoneme in the speech information, and for the display device 120. Output (step S108).
[0041]
Subsequently, the learner 2 performs pronunciation practice for each syllable or phoneme he / she is not good at based on the evaluation score for each word or phoneme (step S110).
[0042]
When it is determined that the pronunciation practice has been completed, the learner 2 gives an instruction as to whether or not to try to pronounce the model syllable again via the input device (keyboard or voice input device) of the personal computer 110 (step S112). If it is instructed to try again, the process returns to step S104. If it is not instructed to try again, the process proceeds to the next step S114.
[0043]
Next, the learner 2 gives an instruction as to whether or not to try pronunciation practice for other model syllables via the input device of the personal computer 110 (step S114). If it is instructed to try, the process returns to step S102, and if not instructed to try, the process ends (step S116).
[0044]
[Other examples of expanded phonographs]
FIG. 5 shows an extended format speech table that enables practice of vowels by adding vowels peculiar to English to Japanese speech tables.
[0045]
In addition, FIG. 6 shows an extended note table for comprehensive practice of vowels and consonants. That is, the extended sound table shown in FIG. 6 is a combination of parts of the sound tables shown in FIGS.
[0046]
Practicing pronunciation of consonants + vowels, consonants + consonants, consonants + consonants + vowels, etc. using the expanded table as shown in Fig. 3 and Figs. By pronouncing as an extension of, it is possible to practice foreign language sounds more efficiently.
[0047]
FIG. 7 is a graph showing the result of pronunciation practice using the above-mentioned sound table.
[0048]
The experiment was conducted with 7 Japanese as subjects, and the RL and consonant pronunciation of each subject was scored on a 100-point scale immediately before and after training using the sound table. As a scoring method, the speech recognition result by the speech recognition unit 114 was compared with the teacher speech. FIG. 7 shows the change in the average point of the experimental results of the seven persons obtained in this way. The time required for training using the sound table is about 15 minutes, and although it is a short training, it is effective for improving the pronunciation ability.
[0049]
In the above description, the configuration as a foreign language learning apparatus has been described. However, the present invention is not limited to such a case, and software for executing the foreign language learning method as described above is recorded. By using this recording medium, this software can be implemented by a personal computer having a voice input / output function.
[0050]
Furthermore, the software for executing the foreign language learning method as described above is not only installed in a personal computer as a recording medium, but also a personal computer having a voice input / output function through an electric communication line such as the Internet. It can also be realized by installing this software on a computer or the like.
[0051]
The embodiment disclosed this time should be considered as illustrative in all points and not restrictive. The scope of the present invention is defined by the terms of the claims, rather than the description above, and is intended to include any modifications within the scope and meaning equivalent to the terms of the claims.
[0052]
【The invention's effect】
As described above, according to the foreign language learning device or the foreign language learning method according to the present invention, a learner who speaks Japanese as a native language can practice pronunciation of a foreign language (for example, English) in the Japanese alphabet. This makes it possible to practice foreign vowels and foreign language consonants more efficiently for Japanese people.
[0053]
Moreover, consonant consonant practice that is difficult for Japanese people to learn in English, etc., can be acquired through such extended pronunciation table pronunciation practice, enabling Japanese to practice foreign language pronunciation more efficiently. It is.
[Brief description of the drawings]
FIG. 1 is a schematic block diagram for explaining the configuration of a foreign language learning apparatus 100 according to the present invention.
FIG. 2 is a sound table for explaining a Japanese sound system.
FIG. 3 is an extended phonetic table for practicing English consonants.
FIG. 4 is a flowchart for explaining a foreign language learning flow realized by foreign language learning apparatus 100;
FIG. 5 is an extended format sound table that allows vowels to be practiced by adding English-specific vowels.
FIG. 6 is an extended sound table for comprehensive practice of vowels and consonants.
FIG. 7 is a graph showing a result of pronunciation practice using a sound table.
[Explanation of symbols]
2 learners, 100 foreign language learning devices, 102 microphones, 104 speakers, 110 microcomputers, 112 speech input / output units, 114 speech recognition units, 116 arithmetic processing units, 118 data storage units, 120 display devices.

Claims (6)

A foreign language learning device,
Storage means for holding the extended phonetic table to be pronounced by the learner and the corresponding model phoneme information;
The extended sound table, the Japanese syllabary table of Japanese, a sound table obtained by adding at least one of specific consonant or vowel foreign language learning target,
Output means for presenting in advance to the learner the syllables to be pronounced in the extended phonogram;
Phoneme separation means for receiving speech information produced by the learner and separating it into phoneme speech information for each phoneme;
Evaluation means for evaluating the degree of matching with the model phonological information for each phonological speech information ;
Comprising a display unit for displaying the determination result of the previous SL evaluation means prior Symbol phoneme basis or every syllable, foreign language learning device. The extended pitch table is
The foreign language learning apparatus according to claim 1, wherein the [r] sound and the [l] sound are added to the vowel array side. A foreign language learning method executed by a computer including a voice input device, a storage device, and an output device ,
In the storage device of the computer, the alphabetical table of Japanese, and a model phoneme information corresponding to the expanded sound table data by adding an at least one of specific consonant or vowel foreign language learning target Storing, and
Presenting the syllable to be pronounced by the learner from the extended sound table by the output device ;
Receiving a speech signal corresponding to a speech produced by a learner from the speech input device, and performing calculation processing by the computer to separate into phoneme speech information for each phoneme;
The computer calculates and evaluates the degree of coincidence with the model phonological information for each phonological speech information;
And a step of displaying through the output device prior Symbol phoneme basis or every syllable the evaluation result, foreign language learning method. The foreign language learning method according to claim 3, wherein the extended sound table is obtained by adding [r] sound and [l] sound to the vowel array side. The storage device of the computer having an audio input and output devices, the alphabetical table of Japanese, and the extended sound table obtained by adding at least one of learning specific consonant or vowel foreign language of the target data Storing corresponding model phonological information;
Presenting the syllable to be pronounced by the learner from the extended sound table by the output device ;
Receiving a speech signal corresponding to the speech produced by the learner from the speech input device, and performing a calculation process for separating the speech signal into phoneme speech information for each phoneme;
Evaluating the degree of matching with the model phonological information for each phonological speech information;
Recording a computer-readable medium storing a program for executing the foreign language learning method on a computer and a step of displaying through the output device prior Symbol phoneme basis or every syllable the evaluation results. The computer-readable recording of a program for causing a computer to execute a foreign language learning method, wherein the extended sound table is obtained by adding [r] sound and [l] sound to a vowel array side. 5. The medium according to 5.

Images