JP2005031604A - English learning system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable even a beginner who learns English to learn colloquial English spoken by natives. <P>SOLUTION: This language learning system fetches a learner's read-aloud speech, calculates the speed ratio and correlativity to native English which is previously held as a reference sound, and provides the native English for the learner at a reproduction speed corresponding to the learner, so the learner is able to learn English with a natural English sound corresponding to the skill of the learner and then efficiently learn pronunciation, a fading-out sound, delicate rhythm, etc. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

Detailed Description of the Invention

Technical field to which the invention belongs

  The present invention relates to a technique related to the extension of English conversation skills of Japanese people, and more particularly to a technique that contributes to improvement of English listening and speaking ability.

  English is a compulsory education, and considering the rate of advancement to university and high school, many people study English for 3 years at junior high school plus 3 years at high school for a total of 6 years, and further to university. . This has been successful for many years in introducing foreign cultures and technologies into our country through book literature. On the other hand, it is said that the growth of Japanese conversation skills is not ridiculous. English education in Japan is reminiscent of reading and writing education, and there is a reflection on the lack of education on the growth of listening and speaking skills. This part depends on the ability of English teachers to listen and speak, and there are many Japanese English teachers, so it is quite different from English spoken by native speakers of English (hereinafter referred to as native English). Most of the classes are taught in different so-called Katakana English. As a result, an English learner could not understand even when spoken in native English, and noticed that it was not understood even though it was spoken with only sentences of English words that he knew for the first time in writing Etc. exist.

  There are schools that take in English classes of teachers whose native language is English, but it is impossible in terms of numbers to make all English classes in Japan teachers whose native language is English. In addition, the current English class content is better in Japanese. The current level of English literacy is gained by Japanese English teachers, and foreign teachers cannot expect such results. That is, for example, it is necessary to have a lot of vocabulary in order to listen to and understand grammar explanations in English in reading and writing education, and it is impossible to become a teacher whose native language is a junior high school English class itself.

  As mentioned above, the majority of English teachers cannot change the current situation of Japanese people. There is a way to use native English recorded as a teaching material instead of Katakana English. However, because the native English sound is too fast compared to the reading speed of the learner, there is a problem that the learner cannot hear well because the pronunciation of English is different, there is a unique rhythm of English, etc. There is a problem in using native English at normal speed as a teaching material, especially for beginners of English.

Problems to be solved by the invention

  The first problem is to provide a system that enables even a beginner to learn native English spoken normally.

The second problem is to provide a system that provides the learner with an index that indicates an extension of the ability of the learner, such as pronunciation, during reading.
There is a problem that it is not possible to clearly grasp the expansion of the ability of the learner when reading aloud. In the method based on the subjectivity of a third party such as a teacher, if the improvement of the learner's ability is subtle, there is a risk that the variation due to the individual difference ability of each teacher will increase, and the learner himself can grasp the growth of ability. First of all, it is necessary to maintain a strong will to maintain learning motivation.

  The third problem is that it is effective to listen and speak at as many occasions as possible to extend the ability to listen and speak, and provide a system that can learn a lot in everyday life.

  The fourth issue is native English, but English in these districts, such as British English, Australian, and US English, has a different pronunciation. In the English-speaking area, such as southern English, there is a difference that the pronunciation, Liz's, etc. are not practically ignorable depending on the district.

  The fifth problem is to provide a system in which the learner can concentrate on learning while minimizing device operation when learning is performed. For example, although it may be effective to reproduce the same part several times at different speeds, it is necessary to operate the device as much, and the problem of being unable to concentrate on learning due to that operation is solved.

Means for solving the problem

  For the first issue, the learner's voice-reading sound is captured, the speed ratio and the degree of correlation with the native English pre-stored as the reference sound are calculated, and the native English is sent to the learner at a playback speed according to the learner. A learning system that provides learning content that matches the learner's ability.

  For the second problem, a learning system that outputs the score information S to the learner based on the correlation value and speed ratio between the learner's reading voice and the reference sound is a reference for the progress of the ability to each learner. Information can be provided, and learners themselves can check if there are daily learning outcomes.

  With respect to the third problem, a learning system in which a voice input unit and a sound generation unit are connected to a system via a communication line, and learning can be freely performed even in a remote place.

  For the fourth task, British English, Australian English, American English, etc., which are characteristic sounds of English-speaking districts, are recorded in advance as multiple standard sounds, and learners respond to their own wishes. A learning system that selects the English of the district from a plurality of standard sounds, and can learn the English sounds of the district desired by the learner.

  A fifth problem (apparatus operation) is a learning system that can store a reference sound reproduction procedure and its reproduction speed as a scenario program in advance, and a learner can minimize apparatus operation during learning.

First Embodiment of the Invention

  This will be described with reference to FIG. When the learner reads out the printed English text, the voice input unit 10 captures the learner's voice reading as input and stores it in the storage unit 20. In the reference sound recording unit 30, native pronunciations having the same contents are stored in advance as reference sounds. The calculation unit 40 performs dynamic frequency analysis for the reference sound of the reference sound recording unit and the sound of the storage unit, and outputs a correlation value r (max) of both dynamic frequency components and a speed ratio coefficient k. The reproduction unit 50 reproduces the reference sound based on the reproduction speed based on the speed ratio coefficient through the sound generation unit 60. Since this is the playback speed according to the student's ability, you can reasonably study subtle pronunciation differences and unique rhythms of native English.

The dynamic frequency analysis performed by the calculation unit 40, the correlation value r (max), and the speed ratio coefficient k will be described in some detail.
In dynamic frequency analysis, as is known in voiceprint analysis, for example, the horizontal axis is the time axis and the vertical axis is the frequency axis, and each coordinate, that is, the intensity of a certain frequency component at a certain time is calculated. It is indicated by a three-dimensional map such as color and shading. If the student's generated sound is slower than the reference sound, the time axis of the dynamic frequency component map is extended compared to the reference sound. If the pitch is high, the three-dimensional map is shifted upward. If the pronunciation itself is different, the 3D map pattern changes.
The three-dimensional map of the sound in the storage unit 20 and the three-dimensional map of the reference sound in the reference sound recording unit 30 are expanded and contracted in the time axis direction and in the frequency axis direction. Find the maximum state. The correlation value at this time is r (max). The degree of expansion and contraction in the time axis direction at this time is the speed ratio coefficient k. The expansion and contraction movement in the frequency direction at this time means the pitch of the voice.
Since the reference sound reproduced based on the speed ratio coefficient k reflects the ability of the learner, it becomes a reference for setting the speed of the sound that can be heard and understood with the least resistance. This allows students who are beginners of English who can only pronounce fluent English to learn the rhythmic characteristics of English pronunciation differences while listening to native English at a speed that suits the students.

  The speed ratio coefficient k and the reproduction speed of the reference sound need not be the same. This is generally due to the fact that hearing ability and reading ability are not identical. For example, when k = about 0.5, that is, when the learner's voice is half the speed of the reference sound, if the speed of the reference sound reproduction is the same half, it is rather too slow, for example, A speed of about 65% may be a listening speed suitable for the learner's ability. Therefore, the learner may determine the relationship between the speed ratio coefficient k and the reproduction speed.

  The voice input has been described on the assumption that the student reads out the type, but instead of reading out the type, the playback unit may play a reference sound at a predetermined speed, and the learner may repeat the pronunciation and use this voice as the voice input. In the case of a format that repeats the standard voice, there is an advantage that the learner can easily carry out the preparation without having to prepare a separate type.

  Based on the correlation calculation result value r (max) and the speed non-coefficient k, the score information S can be shown to the learner. As a result, an objective score can be obtained for each individual as an indication of progress due to the progress of learning, so that a sense of achievement can be given to the learner, encouragement of learning, and easy maintenance of learning motivation. The score information S may be a 100-point perfect score method or may be converted into a grade.

  The above-mentioned reference sounds are recorded as multiple reference sounds, such as British English, Australian English, American English, New York, Western, etc. You may make it select. Thereby, since the English sound of the district which a learner desires can be learned, the English sound of the English-speaking area which a learner desires can be learned more.

Correlation values between the plurality of reference sounds and the learner's reading voice may be calculated as the correlation coefficient calculation target, and the district name having the maximum correlation value may be output to the learner. As a result, the learner's pronunciation is the most similar English in which district, and learning can be continued happily and the learning motivation can be easily maintained.

  The playback unit reproduces the reference sound in advance with the most convenient content for each learner, such as a 100% native English sound followed by an 80% repeated sound followed by a 65% repeated sound. You may make it record as a program of a procedure. As a result, the study can proceed with only the minimum operation of the device, so that the user is not conscious of the operation of the device and can concentrate more on the study.

  The native English reproduced based on the program of the reference sound reproduction procedure may be recorded on a medium, and the medium may be reproduced and learned. Thereby, the learning system environment of this invention is realizable only with a sound reproduction apparatus and the said medium. Some sound reproducing devices are easy to operate, small and light, and have excellent portability. Therefore, a learning environment with extremely high portability can be realized.

Second Embodiment of the Invention

Another embodiment of a highly portable learning environment is shown. As shown in FIG. 2, a reference sound may be received as a model voice through a terminal 1 having a voice input unit and a sound reproduction unit 60 and a communication line such as a wireless telephone line or the Internet, and it may be repeated. The terminal 1 may be a mobile phone, for example. In this case, it is convenient to repeat reading instead of reading aloud. For example, learning is possible with only a mobile phone, and an extremely portable learning system can be realized.
As described above, learning can be performed anywhere if there is a bidirectional communication terminal having a voice transmission / reception function even in a remote place.

  As explained above in English learning, the learning system of the present invention can remove the premature point of the sound to be learned in foreign language learning, and more clearly learn differences in pronunciation, intonation, rhythm, etc. It can be easily imagined that it can be applied to learning foreign languages other than English.

The invention's effect

As described above, according to the English learning system of the first invention, it is possible to learn with a native English sound having a playback speed according to the learner's ability, so that learning of differences in pronunciation, subtle rhythms, etc. can be efficiently performed. Yes.
According to the English language learning system of the second invention, the score information S corresponding to the reading sound is read to the learner, so that the progress of the learner's ability can be grasped at any time, and the information that encourages the learner accordingly. Can be provided.
According to the English learning system of the third invention, it is possible to learn freely and easily when the learner desires even in a remote place.
According to the English learning system of the fourth aspect of the invention, the learner can select a desired English-speaking region and learn the English sound of that region, so that efficient learning can be realized.
According to the English language learning system of the fifth aspect of the invention, since device operation can be set in advance, it is possible to concentrate on learning by releasing from troublesome device operation and to realize efficient learning.
According to the English learning system of the sixth aspect of the invention, desired learning can be performed only with the recording medium and its playback device, so that it is excellent in portability and can be efficiently learned.

Functional block diagram for explaining the first embodiment of the present invention Functional block diagram illustrating a second embodiment of the present invention

Explanation of symbols

1. Terminal 10. Voice input unit 20. Storage unit 30. Reference sound recording unit 40. Calculation unit 50. Playback unit 60. Sound generator

Claims (6)

A voice input unit 10 that captures a learner's utterance, a storage unit 20 that stores the utterance, a reference sound recording unit 30 that has native English as a reference in advance, a voicing sound of the storage, and the above A calculation unit 40 that calculates the reference sound of the reference sound recording unit, a reproduction unit 50, and a sound generation unit 60 that outputs audio information from the reproduction unit;
The arithmetic unit 40 has a function of performing a dynamic frequency analysis of both the sounds and outputting a correlation value r (max) and a speed ratio coefficient k that maximize the degree of correlation between the dynamic frequency components, and the reproducing unit. 50 is an English learning system for reproducing the reference sound at a reproduction speed corresponding to the speed ratio coefficient k.   2. The English language according to claim 1, wherein the calculation unit 40 has a function of calculating and outputting the score information S of the voice of the storage unit based on the correlation value r (max) and the speed ratio coefficient k. Learning system.   The English learning system according to claim 1, wherein the voice input unit (10) and the sound generation unit (60) are bidirectionally coupled to a remote place via a communication line.   The English learning system according to claim 1, wherein the reference sound recording unit stores a plurality of reference sounds having characteristics of sounds generated in a plurality of regions.   The English learning system according to claim 1, wherein the playback unit (50) stores the reference sound setting and playback procedure as a scenario program, and plays back the reference sound based on the scenario program.   6. The English learning system according to claim 5, wherein a sound obtained by reproducing the reference sound by the reproducing unit based on the scenario program is recorded in advance on a recording medium, and learning is performed using the storage medium.

Images