WO2000045288A2

WO2000045288A2 - Computerized translating apparatus

Info

Publication number: WO2000045288A2
Application number: PCT/IL2000/000060
Authority: WO
Inventors: Jacob Fromer
Original assignee: Jacob Fromer
Priority date: 1999-01-31
Filing date: 2000-01-30
Publication date: 2000-08-03
Also published as: IL128295A0; WO2000045288A3; AU2127000A; IL128295A; EP1149349A2; JP2002536720A; CN1339133A

Abstract

A computerized dictionary/translating apparatus is disclosed, which comprises an additional screen (G) and suitable software for displaying, besides the textual form of the retrieved/translated word, also the mouth articulation of the word when verbally pronounced.

Description

COMPUTERIZED TRANSLATING APPARATUS BACKGROUND OF THE INVENTION

The present invention generally relates to computerized translators. In the art of computerized self-teaching apparatus there are presently available the kind known as "Electronic Dictionaries". These devices comprise a keyboard and a display screen. Words entered in one given language will be displayed on the screen translated into another language according to the programming of the device. More advanced apparatuses include an additional audio feature of producing, through suitable electronic circuitry and built-in speaker, the spoken sounds of the translated word.

This important assistance to users is insufficient, in the sense that it is still missing a guidance as how to exactly pronounce the word, namely the visual presentation of the mouth and related organs (lips and tongue). It is thus the object of the present invention to provide a computerized translator capable of displaying the actual pronunciation of the translated words.

It is a further object of the present invention to utilize for that purpose a database of basic phonetic syllables of which all words in the translated language are composed.

It is a still further object of the present invention to display the pronunciation of the translated words in a dynamic streamlined fashion, using the morphing technique. SUMMARY OF THE INVENTION

Thus provided according to the present invention is a computerized

dictionary apparatus, comprising a first database file of words in a given

source and target languages, and a second database file of textual

pronunciation keys of the basic phonetic syllables relating to the words of

the said first database file of words characterized by a third database file of

sequential pictorial mouth articulation images for each of the said basic

phonetic syllables, means for the suitable selection of the source and target languages, means for inputting a selected word included in the first

database file, means for displaying the selected word in a textual form,

means for locating the selected word in the first database file of words in a

given source language, means for selecting the sequential pictorial mouth

articulation images relating to the phonetic syllables of the textual key

pronunciations and means for displaying the sequential pictorial mouth

articulation images of the selected word in differential and combined

succession.

BRIEF DESCRIPTION OF THE DRAWINGS

These and additional objects, advantages and features of the present

invention will be more clearly understood in the light of the ensuing description

of preferred embodiments thereof, given by way of example only with reference

to the accompanying drawings, wherein-

Fig. 1 illustrates the general design of an apparatus featuring the

characteristics of the present invention; Fig. 2 is a block diagram of the apparatus sub-systems; Fig. 3 is a flow-chart of the apparatus operations; Fig. 4 is an example of use of the method as applied to a specific word; and Fig. 5 illustrates the morphing process.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS As schematically shown in Fig. 1 the proposed apparatus generally designated A is similar to any known electronic translator; it necessarily comprises keyboard B for entering words in the language of origin, suitable keys C for selecting indexed library databases, source language, etc.; and other functional control buttons D, e.g., target language selector; a first LQ screen E for displaying the entered as well as translated words; and, as required for the implementation of the present invention - a second LQ screen G on which the movements of the speech organs are displayed, as will be described in greater detail hereinbelow, and speaker F.

Obviously, the design of the apparatus may vary in many respects, as in fact there are many types of electronic dictionaries available on the market.

It should be further mentioned that the audio function, represented by the speaker F is not essential to the implementation of the invention, although the incorporation thereof seems natural and preferable in the present context.

As seen in Fig. 2, the apparatus comprises the following main software applications: User's interface 10, data processor 12, ROM memory 14 and output interface 16.

In more detail, the user's interface comprises data input unit 20 for converting the data inputted by keyboard B into a textual form, possibly including spell checker function; and a database file selector 22 for interpreting commands entered by language selector buttons C and D (Fig. 1 ) to select source or target languages from vocabulary database files 24 of the ROM memory application 14. The data processor 12 comprises data searching engine 26, applicable for searching database files according to index or file ID-field names, or for matching records of any two or more database files by index or common ID-fields, using data integrator 28.

The ROM memory application 14 comprises the above mentioned vocabulary database files 24. It contains a vocabulary of words in a given language available for selection by index or by alphabetical order; the index of each database is linked to other databases, to create mutual matching.

Word pronunciation database files 30 contains records of textual phonetic pronunciation, composed of basic syllables for every word included in the vocabulary.

Images of mouth articulation database files 32 contains records of sequential image sets of all basic phonetic syllables.

Key-point sets of mouth images database files 34 contain records of all key-points sets of mouth images, outlining the first and the last image of each syllable.

Image morphing generator 36 is also included for creating transient images, preferably by buffer 38 (see below).

The output interface application 16 is used for presenting the translation results to the user in a visual form and, optionally, also in audio form (speaker driver 40), and comprises text display driver 42 and mouth articulation display driver 44.

The operation of the apparatus according to the present invention is schematically represented in Fig. 3.

The first step of the user is to select out of all available languages programmed into the apparatus A the language that he/she determined as the source or origin language, and the desired target language. The database file selector 22 selects the matching source vocabulary database file from database files 24.

Let us assume that the source language chosen is French, and the word "BONJOUR" is requested to be translated into English.

Data input unit 20 is applied for inputting the words by keyboard B. The selected word is displayed on screen E using text display driver 44, and is located by the data search engine 26 in the selected source language vocabulary database files 24. The user selects the target language for the English translation. Let us further assume that the word "HELLO" has been retrieved as the translation of "BONJOUR". Database file selector 22 selects the English language vocabulary database file from vocabulary database files 24, using the data integrator 28 for matching between source and the target vocabulary database files. The translated word is being now matched by data integrator 28 with the corresponding textual pronunciation keys stored in database files 30 containing the basic phonetic syllables of which every word is composed. Thus, the word "HELLO" is composed of two basic phonetic syllables: h and I , as represented in Fig. 4.

As further illustrated in the flow-chart of Fig. 3, each one of the basic syllables is matched with the associated set of mouth articulation images, as stored in the mouth articulation database files 32 (schematically presented as a series of ellipses).

Now, by displaying on screen G (Fig. 1) the relevant sets of mouth articulation images, (employing visual display driver 42) a visual presentation of the word pronunciation is presented, concurrently with the audio production of the word. It is advisable that the duration of each image appearance will be controllable, and also that each syllable separately could be repeated as many times as desired.

It will, however, be noted that the sequential presentation of mouth articulation image sets may not yield a streamlined pictorial presentation, but rather appear as a step-wise animation.

To rectify that, and according to an additional feature of the present invention, the apparatus is provided with an image morphing generator 36, whose function is to attain full, smooth and realistic demonstration of word pronunciation. By this image morphing function (which is known per-se and need not to be described in greater detail), transient or dynamically changing images are created. These transient images are dynamically and intermediately

inserted in-between sets of two successive syllables, to achieve streamlined pictorial presentation.

The image morphing function enables to create sequential series of transient images, from the last image of the first syllable to the first image of the

second syllable.

A simple example of the morphing process application is schematically

depicted in Fig. 5. The new images are stored in buffer 38.

Fig. 5a schematically shows the last mouth image of the first syllable,

outlined by selected key-points A1 - A6, whereas points A1'" - A6'" in Fig. 5d

outline the last mouth image of the second (and last) syllable. The selected

key-points are retrieved from key-point sets of mouth images database file 34.

The product images of the morphing process are shown in Figs. 5b and

Fig 5c, namely being slightly shifted at every step to reach the position of the

target image in Fig. 5d.

The application of the morphing process is integrated into the flow-chart

of Fig. 3 (last two rows).

Using the visual display unit 42 to display on screen G the new transient

images in-between the syllable images, gives the user a vivid visual

presentation of the word pronunciation.

It will be readily understood that the apparatus may be designed to

comprise a ready-made database file of transient images matching all

possibilities of filling in-between any couple of basic syllables. This database file will replace the morphing function. However, this option will require an excessively large database file, and therefore rejected for practical reasons. According to yet another option, the apparatus may comprise a database file of sequential image sets demonstrating the pronunciation of all complete word belonging to the selected language. This again will dictate an unreasonably large database and therefore disadvantageous compared with the other possibilities as described in detail hereinabove.

While the above description contains many specificities, these should not be construed as limitations on the scope of the invention, but rather as exemplification of the preferred embodiments. Those skilled in the art will envision other possible variations that are within its scope. Accordingly, the scope of the invention should be determined not by the embodiment illustrated, but by the appended claims and their legal equivalents.

Claims

WHAT IS CLAIMED IS:

1. A computerized dictionary apparatus, comprising a first database file (24) of words in a given source and target languages, and a second database file (30) of textual pronunciation keys of the basic phonetic syllables relating to the words of the said first database file (24) of words c h a r a c t e r i z e d by-

- a third database file (32) of sequential pictorial mouth articulation images for each of the said basic phonetic syllables;

- means (22) for the suitable selection of the source and target languages;

- means (20) for inputting a selected word included in the first database file (24);

- means (44) for displaying the selected word in a textual form;

- means (26) for locating the selected word in the first database file (24) of words in a given source language;

- means (32) for selecting the sequential pictorial mouth articulation images relating to the phonetic syllables of the textual key pronunciations; and -means (42) for displaying the sequential pictorial mouth articulation images of the selected word in differential and combined succession.

2. The apparatus as claimed in Claim 1 , further c h a r a c t e r i z e d by-

- a forth database (34) of selected key-point sets outlining the said images for each first and last sequential pictorial articulations of the phonetic syllables;

- means (36) for creating the transient pictorial articulation images between two phonetic syllables using the key-point set of the last pictorial articulation of the first phonetic syllable and key-point set of the first pictorial articulation of the second phonetic syllable, and so forth; and -means for synchronically displaying the transient pictorial articulations in-between of the respective pictorial articulations of the phonetic syllables.

3. The apparatus as claimed in Claim 1 , further comprising means (40) for audio reproduction of said textual word pronunciation and said basic phonetic syllables, in synchronization with the visual displaying of said basic phonetic syllables involving a morphing images processing.

4. The apparatus as claimed in Claim 1 , wherein the inputting data means (20) comprise a keyboard (B).

5. The apparatus as claimed in Claim 1 , wherein inputting the selected word in the first source language becomes associated with corresponding word in the given second target language by means of data search engine (26).

6. The apparatus as claimed in Claim 1 , further c h a r a c t e r i z e d by -

- a database of transient pictorial articulations necessary for filling-in of the intervals between successive pictorial articulations of the basic phonetic syllables; - means for locating the transient pictorial articulations;

- means for synchronically displaying the transient pictorial articulations in-between of the respective pictorial mouth articulation images of phonetic syllables; and

- means for combining of the aforesaid pictorial mouth articulation images with corresponding audio files and their controllable simultaneous differential and/or combined synchronical playback.

7. The apparatus as claimed in Claims 14, further c h a r a c t e r i z e d by means for automatic repetitions of the differential and combined visual mouth articulation images and audio differential and combined playback of any inputted word from the first source language database file and /or second, third etc, target language database files.