Note: Descriptions are shown in the official language in which they were submitted.
~Z~94
The present invention relates to a portable translator
device which translates words in a first language into words
in a second language.
In the past, travelers unfamiliar with local language
would refer to a pocket dictionary, which provided an abbreviated
dictionary of translated words. Computer operated translators
have been devised for use by professional translators. However,
the expense, bulk, and overly extensive vocabulary make it
unavailable and impractical for the ordinary traveler.
Advances in mini-electronics for portable calculators
and advances in electronic speech synthesis have generated new
and adaptive electronic components. The problems of using such
components in an arrangement and size suitable for a portable
translator has not been solved.
It is an object of this invention to provide a portable
translator device that is small in size for convenience, in-
expensive and that can provide an synthesized, oral pronunciation
of the word translated useful to a traveler.
This invention provides a translator device for trans-
lating words in a first language to words in a second languageand conversely, comprising a combination of a microcomponent
assembly for digital storage and processing of data, and a pre-
defined data base arrangement characterization in a self-contain-
ed portable hand-holdable unit for selective visual and audible
output of words in said first language and in said second language
having: word input means and function input means for manually
inputting words and commands for conversion to electronic signals;
an interface circuit for digitally generating and temporarily
storing equivalent processable word information and instruction
information from the electronic signals for translation of words
that have been input through the word input means; a central
processor unit with an associate control program for directlng
~ I
~Z9~94
operations under instruction from the control program; a clocking
means operating in conjunction with said central processor unit
for use in timing the coordination of said operations; control-
ler means electrically connected to said interface circuit and
to said central processor unit for receiving and routing word
information and instruction information originated through said
word and function input means to unit components of the trans-
lator device under direction of said central processor unit; a
memory wherein said control program is stored for use by said
central processor unit in directing translation and generating
output words, and having a stored data base, said data base
comprising: a predetermined arrangement of word information of
a first language together with a predetermined arrangement of
associated speech synthesis code information for assembling-
speech synthesis audio information for generating audio output
words corresponding to said word information of said first
language; a predetermined arrangement of word information of a
second language, together with a predetermined arrangement of
associated speech synthesis code information for assembling
speech synthesis audio information for generating audio output
words corresponding to said word information of said second
language, said word information of said second language compris-
ing language translations of said word information of said first
language; code data associated with said word information of
said first language and said second language, for use by said
control program in locating word information in one language
corresponding to words input through said word input means and
locating the corresponding word information and speech synthesis
code information in the other language; and a predetermined
arrangement of speech synthesis audio information for assembly
in generating audio output words; output means under operation
of said controller means by direction of said central processor
. .
1~29~94
unit for producing visual output words from located word infor-
mation from said data base and or generating audio output words
from assembled speech synthesis audio information from said
data base upon manual input of input words and selected commands
for translation of an input wordi and a power supply means for
powering the electrical components of said translator device.
The invention will now be described in greater detail
with reference to the accompanying drawings in which:
Figure 1 is a perspective view of the preferred single
unit electronic translator device;
Figure 2 is a block diagram of the micro-computer
system in the device of Figure l;
Figure 3(a) is a schematic illustration of the storage
arrangement in the read only memory;
Figure 3(b) is an alternate arrangement of the storage
of 3(a);
Figure 4(a) is a flow chart illustration of the
operation;
Figure 4(b) is a flow chart illustration of a sub-
sequence; and
-2a-
~Z9i94
(3)
Figure 5 is an alternate embodiment o~ the elec-
tronic translator device in two line connected components.
Referring to Figure 1, the preferred embodiment
of the electronic translator device, designated generally by
the reference numeral 10, is shown. The device 10 has an
enclosing case 12 with a front panel 14. The front panel
has a dual display 18 at the top of the panel.
The keyboard is arranged primarily around an alpha-
betical set 20 of keys in typewriter configuration to facil-
itate input as is conventional in keypunch and other terminals.
A plurality of function keys 21 allows various operations to
be performed and increases the versatility of the translator.
The translator 10 is devised to allow input of
a word or words in a first language and generate an output
of a word or words in a second language which comprise an
automatic translation of the words in the first language.
The words of the first language are displayed in the first
line 22 of the display 18 and the translated words are dis-
played on the second line 24 of the display.
Furthermore, the translation device includes a voice
synthesizer which generates an oral pronunciation of the trans-
lated word through a speaker 26 having a volume control 27.
This is an immense aid to those who are unfamiliar with a
foreign language and need immediate aid for speaking.
A limited memory included in the device stores a
phrase or short sentence by stringing together several words
which may be translated and displayed, and if desired,
orally spoken. If the information to be displayed exceeds
the display capacity, a lamp 28 will light indîcating the dis-
play is incomplete. Depressing a display key 29 will cause
Z9~94
display of the xemainin~ data.
Certain common words and short phrases are
directly translated. These, however, are limited in num-
ber, requiring an individual key Eor each word or phrase.
The alphabetical set 20 for word input is trans-
~ormed to a short word set by depressing a function key
30 labeled "2nd". This key shifts the alphabetical re-
gister to a common word register for the words or phrases
indicated below each key. An individual can translate
directly these frequently used words or phrases without
inputting individual letters to s~pell the desired word
or phrase.
A third register reached by depressing the "3rd'
key 25 translates numbers, that is, displays the numbers
in written form in the foreign language.
As noted, a principal function of the device is
to provide an oral translation of the input word or words.
After activating the device by the on-off switch 31 on the
side of the device and selecting the direction of transla-
tion by slide 32, along the edge of the device, speech cir-
cuits can be act~yated for use by a sw~tch 33. When a word
has been keyed in ana a trans-lation, includin~ oral~ is
desired, a com~ned translation ana speech key 34 i~s de~
pressed. The oral synthes-is of the word can be repeated
by depress~in~ the repeat key 35. If si~ply~ a translat~on
is requ~red, a translation key 36 ~s~depress-ed and the oral
translation i5 om~tted. If no oral translations are con-
templatd~ the activating sw~tch 33 for the oral synthe-sizing
circu~ts i`s-ma~ntained in the "of~" post~on to savè the
power supply.
A set o~ memory keys comprising a memory store key
li29194
(5)
38, a memory reproduce key 40 and a memory clear key 42
allows temporary storage of words, reproduction of the
stored words in a phrase or short sentence and clearing of
the memory. Thus, several words can be stored and simultan-
eously displayed to allow the user to assemble a phrase or
short sentence. This capability is naturally limited by
the capacity of the display which can conveniently accomo-
date about forty characters. Display key 29 as noted is
used to display the remaining data in excess of the display
capacity.
A tense key 44 and a plural key 46 will display
different tenses in se~uence on each key depression or
respectively the plural of the input word, if desired. This
is particularly help~ul if the ~rds have irregular tenses
or plurals. For saving in memory space, such tense and plural
selections are provided only if irregular. Signal lamps 47
indicate such are available.
A set 48 of accent keys, are included to correctly
input words in various languages where accent marks or an
apostrophe are necessary.
For the particular device of Figure 1, an inverse
display key 50 is included which permits a word that is trans-
lated from a first language to a second language to be retrans-
lated from the second language to the first thereby providing
a check on the meaning. This aids in ascertaining whether
the proper word, if multiple translated words are offered,
has been correctly selected. Other keys such as a common
expression key 51, which by lamp 52 signals when an input
llZ9~94 `
(6)
word is used in one or more common expressions stored in
memory and a form of speech key 49, which allows selection
of the desired form of speech, i.e. noun, adjective, etc.
when available, complete the conte~plated special operational
functions. A clear key 53 allows the entry to be cleared
in a conventional manner.
~ efore detailing the practical manner of using the
translator device, a consideration of the block diagram of
Figure 2 will provide an understanding of the internal
operation of the device. The primary components are integrated
circuits which, for example, correspond to existing shelf
item circuits manufactured by Intel Corporation as set forth
in the chart included hereinafter. The resultant assemblage
is in effect a special application microprocessor system
using a circuit configuration that constitutes a simple
general purpose digital computer.
The microprocessor system consists of a central
processor unit 54 with a clock generator 55 and controller
circuits 56, random access memories 58, read only memories
60 and several input/output devices. The input~oùtput devices
include a keyboard 62 and a visual display or character display
64, as discussed in detail herebefore, a speech synthesizer
designated generally by reference numeral 66 and an optional
printer 68. The central processor 54 co~municates with the
memories and input/output devices via the data 70, control
72 and address 74 busses. Special interface circuits 76,
78, 80, are used to interface the bus lines to the associated
input/output devices.
~lZ9~94
(7)
The keyboard 62 is a switch matrix which connects
to the keyboard/display interface circuit 76 via scan lines
82 and data lines 84 and 86. Data from each key stroke are
encoded and temporarily stored by the interface circuit 76
as schematically illustrated in Figure 2, and an interrupt
request is automatically sent to the central processing unit,
the CPU, 54. The CPU 54 xesponds by reading the data from
the keyboard interface 76 via the data bus 70. The CPU
54 then temporarily writes the encoded letters to the random
access memory 58 for later use and also writes the data to
the display interface 76. The display interface 76 gener-
ates scan and data signals for the visual character display
64.
Program instructions which control the operation
of the CPU 54 a~e stored in the read only memory 60. The
read only memory 60 also contains the spelling and trans-
lation instructions for the dictionary vocabulary. As a
word is spelled into the keyboard 62, the CPU 54 searches
the contents of the read only memory 60 to locate the word.
Associated with each vocabulary word in the read only memory 60
are a set of translation instructions which contain the memory
address of the translated word. These instructions are used
by the CPU to find the spelling of the translated word in
the read only memory and transmit it via the display inter-
face 76 to the character display 64.
The optional printer 68 is able to produce a perm-
anent record of the information appearing on the character
display 64. The peripheral interface circuit 80 translates
`- llZ9~94
(8)
printer instructions received from the CPU 54 into appro-
priate printer commands. The printer is expected to comprise
an auxilliary plug in unit.
The speech synthesizer 66 includes a peripheral
interface 78, a controller 88 and a speech data read only
memory 90 which stores phonemes and synthesis information suf-
ficient to construct speech wave forms that are amplified and
filtered in an amplifier 92 which annunciates words in both
languages through speaker 94. The address codes for speech
vocabulary are sent from CP~ 54 to the speech controller 88
via the peripheral interface 78. The speech controller 88
contains the algorithm for synthesizing speech from data
stored in the speech memory 90. Speech waveforms produced
in the speech controller 88 go through the audio filter and
amplifier 92 to a loud speaker 94.
In addition to letters, the keyboard has several
command keys. These commands are read by the CPU 54 via the
keyboard interface 76 and are executed according to program
instructions stored in the read only memory 60. For example,
using the memory store keys words and phrases can be tempor-
~ily stored in the random access memory 58 for recall on
command.
ThecfQllowing suggested components, some in
multiples, may be used with appropriate software to accom-
plish the above described functions.
9~94
(9)
NO. ITEM COMPONENT IDENTIFICATION
_~ __ _
~,,, ~
54 Central processor unit 1977 Intel Cat. No. 8080A
Clock generator 1977 Intel Cat. No. 8224
56 System controller 1977 Intel Cat. No. 8228
58 Random access memory 1977 Intel Cat. No. 8111
Read only memory 1977 Intel Cat. No. 8708
76 Keyboard/display interface 1977 Intel Cat. No. 8279
78 Peripheral interface 1977 Intel Cat. No. 8255A
Peripheral interface 1977 Intel Cat. No. 8255A
88 Speech controller r
~ Telesensory Systems, Inc.
90 Speech data read only ~I.C. chip S16031-A
memory
Referring to Figure 3(a), the schematic illus-
tration provides a general scheme for arranging storage
of the first and second language vocabularies in the read
only memory 60 of the mircroprocessor.
The arrangement is based on a root word ordering
of the vocabulary. Each block essentially represents one
or more storage bytes and contains a coding symbol, a word
letter or an address code. The example of the root word
ordering in Figure 3(a) is typical of the coding for the
vocabularies of both languages.
The first byte 100 in a coding series indicates
whether the following stored word is a new word, a root word
and/or a continuation of a root word, and additionally
includes coding indicating the part of speech, i.e. a noun,
~rcllQ r~
A
li29194
(10)
verb, adjective, etc., correlative to the translationswhether the word is a key word for phrases, and/or is an
irregular verb or has an irregular plural. A root word is
a word that forms the beginning of a plurality of other words.
The next byte 101 indicates the number o~ bytes
until the beginning of the next continuation of a root word,
which enables a word search to skip the intervening bytes,
and the number of letters in the word.
The following bytes 102 comprise the letters of the
word stored.
Subsequent to the letters of the word are bytes
103 for the address for speech synthesis memory.
Following are three groups of bytes 105, 106 and
107 which comprise the address locations of the corresponding
three definitional words of the example in the second language.
Following these addresses are address bytes 114
for one or more common expression phrases if such include
the word stored as a key word or the word stored is a key
word though not included in the phrase.
This sequence is repeated with minor variations.
For example, the letters stored can comprise both a continu-
ation of a root word and itself a root word 115.
Basically, the following chart indicates the appor-
tioning of storage.
94
(11)
MAIN PROGRAM
VOCABULARY - FIRST LANGUAGE
VOCABULARY - SECOND LANGUAGE
PHRASE LISTINGS
PHRASE LISTINGS
The preferred ordering of storage is based on an
indexing system similar to that of an ordinary dictionary
with addressing for translations to the second language.
The translations are stored in a separate section of the
memory with a similar arrangement of addressing for the first
language. The root word method compresses the size of the
storage and allows the search for a word to skip over contin-
uat~ons of a root word if the root word is not part of the
word sought. If vocabulary size is large, the CP~I can check
what the first letter is and jump immediately to the section
where the first letter starts.
Other storage arrangements may be advantageous.
For example, instead of storing words alphabetically, the
storage can be done by order of frequency of word usage, i.e.
words most frequently used being stored in the front part of
the storage memory. Alternately the translation word in the
second language can be placed adjacent the first language
word in conventional dictionary style. As shown in Figure
3(b), the letters of the translation word in the second language
120 and the speech data 121 for oral synthesis of the trans-
lation can be located adjacent the first language word. The
~2~i94
(12)
addresses of alternate translations 122 and 123 follow together
with the addresses of phrases in the first language 124, 126
and their respective translations 125 and 127. This would
reduce the storage required for a substantial portion of
the cross addressing. However, the storage of first language
words having multiple translations which in turn have mul-
tiple in reverse translations would be difficult and confusing
to arrange in a data base.
The address codes for stored words also include
necessary referral addresses to the speech data read only
memory to enable coding for phoneme assembly in the word
synthesizer. In such instance, the address o~ the phoneme
code in the read only memory can be used to locate the address
in the speech data read only memory, the coding that is
necessary to string together the basic phonemes stored in
digital form in the speech data read only memory.
The divided storage systems for the first language
data base, the second language data base, the common expres-
sion or phrase data base, and the speech synthesis data base
may require a somewhat greater storage capacity than alter-
nate systems. For example, the integrated system shown in
Figure ~(b) can minimize cross addressing by locating the
translation adjacent the base word. However, the divided
storage system described in Figure 3(a) is a preferred
arrangement for systematically establishing the data bank and
for allowing reprogramming for different language pairings.
Referring now to the flow chart of Figures 4(a)
and 4(b) the operation of the translator device can be
described with greater clarity in conjunction with the pre-
vious figures of the drawings.
~Z9194
(13)
In Figure 4(a) the functioning of the translator
commences on activating the device and entering letters
via the keyboard 16. From this start 130 in the flow
chart of Figure 4(a) the CPU 54 reads the letters from the
key board in a read step 132 as they are entered. In a
subsequent store step 134 the CPU stores the letters and
subsequent incoming letters into the random access memory
58, and also displays the letters in the display 18 such
that the user can verify his entry. In a search step 136
the CPU institutes a search of the read only memory 60
using comparer operations to match the incoming letters to
the initial letters and root words of the words stored in
the vocabularly of the read only memory.
In a decision step 138 it is determined if the
letters received are present in proper sequence in at
least the first of a series of stored vocabular~ words.
If not then additional letters input will not cure the
deficient comparison and the display 18 will immediately
flash an "unknown" and display the word from vocabulary
with the closest spelling as indicated in the termination
step 140.
If yes, there is a subsequent decision step 142
to determine if there are more letters. If there are more
letters the loop is continued until there are no more let-
ters. The indicator determining that there are no more
letters is a request to translate by depressing the trans-
lation key 36, or the common expression key 51.
,~;'
~Z~1~'L94
After determining that there are no more letters,
a comparison is made in a comparison step 144 whether the
input word is identical in full to a vocabulary word. If
not, the "unknown" termination 140 is activated with a
display of the closest word stored. If yes, it is defined
in a decision step 146 whether a phrase is wanted, and,
if yes, an automatic determination step 148 defines whether
a phrase is available. If yes, in the terminating step
150, the phrase or phrases are retrieved and displayed
with their translations in the display. If more than one
phrase is available the phrases are sequenced by repeating
the pressing of the common expression key.
If no phrase is available and it is wanted at
this time by selection of the phrase key 51 in deference
to one of the translation keys, a "no common expression"
signal is flashed in the display in step 152.
If in user decision step 146 it is decided that
no phrase is wanted, it is nevertheless automatically
determined if a phrase is available in determination step
154. If yes, a light 52 adjacent the common expression
key on the keyboard is activated, box 156, during the fol-
lowing sequences to indicate that there is one or more com-
mon expressions incorporating as a key word, the word that
is desired to be translated. At any time during or after
the translation while the input word is still being re-
tained, the common expression key can be depressed,
causing retrieval of one or more phrases including the
input word or for which the input word is a key word.
~,,
~l~Z9~94
.
_ (15)
If no phrase is available, the indicator light
remains dark and the sequence goes immediately to a trans-
lation step 158.
If the word to be translated has more than one
form of speech available for translation, i.e. adjective,
noun, verb, etc. and more than one translation, the forms
available and the total number of the translations avail-
able are indicated by symbols. The user selects the form
of speech by an order selection key 49 which sequentially
shifts the order of the symbols representing the forms
of speech. The one farthest to the left, for example, the
"n", in the upper display 22 of Figure 1, can be rotated
to the end and the next in order shifted to the prime
position by this key to short-cut going through all the
translations.
In the flow chart at step 158, the display in-
cludes the input word, its forms of speech, the number of
alternate translations and a translation. The translation
corresponds to the form of speech indicated in the prime
left position, and unless shifted by the translation key,
initially comprises the first translation in that form of
speech.
First there is an inquiry whether the form of
speech is accepted in decision step 160. If no, there is
a shift to the next form at box 162. The display will now
show the next in order form symbol in the prime left posi-
tion together with the first, or only, translation for
that form. If the form is accepted, the translation is
questioned in a similar loop beginning with decision step
.~
- 1~29~94
(16)
164 and shift box 166 where the next in order translation
is provided and continuing until an acceptable or at
least temporarily acceptable translation is located. This
is accomplished by repeated depressing of the translation
key 36.
An inverse translation decision 168, if desired,
displays the inverse translations of the translated words,
i.e. provides all the translations in the first language
of the accepted translation of the second language as a
check on the meaning of the conditionally accepted trans-
lation at box 170. This is accomplished by depressing the
inverse translation key 50. In decision step 172, if it
is found that the accepted translation from this check
was not the one desired the next translation can be
ordered by the translation shift box 166 in the manner
noted.
If the inverse translation affirms the correct-
ness of the translation, the sequence proceeds to the
speech decision step 173. If speech is desired, by depres-
sign the speech translation key 34, speech is synthesized
at box 174. It can be repeated by speech repeat step 176,
representing the repeat key 35, which loops again through
the synthesis. An alternate translation decision can again
be made at decision step 178, which-if desired returns to
the translation shift box 166. If not, the user has the
choice of electing a phrase, if the phrase light is lit
at decision step 180 which goes to the phrase sequence,
here indicated by box 182, a repeat of box 150. If not
the sequence terminates at end 184.
i~2g~g4
(17)
The phrase sequence is similar to the transla-
tion sequence. The phrases are stored in a somewhat dif-
ferent manner as general vocabulary words. sriefly, use-
ful short phrases are stored in a separate section of the
read only memory. The phrases are stored not as spelled
words but as a sequence of address pointers which
correspond to words in the vocabulary memory. This organ-
ization reduces the memory space re~uired for phrases.
Each!phrase is stored as a block consisting of a length
code followed by address pointers. The phrase is called
us by giving the address location of its block beginning.
"Key words" are used for locating a phrase.
Each "key word" may correspond to several phrases. For
example, "taxi" might refer to phrases which include the
word taxi explicitly but might also be a "key word" for
the phrase "how much is the fare?n. In order to keep the
vocabulary memory simple, the phrase addresses are not
located in the respective "key word" memory blocks. In-
stead, the "key word" memory block contains a single
address pointer for phrases which locates an intermediate
list of phrase addresses corresponding to the particular
"key". These intermediate addresses can then be called in
order to construct the actual phrases.
In no, user can proceed to step 194. A speech
synthesis choice is available at decision step 194. If it
is not utilized, the program terminates at end 196; if it
is utilized, the speech is synthesized at box 198. A re-
peat decision step 200 is provided which loops to a repeat
of the synthesis if desired. If not, the sequent terminates
at end 202.
``` ~lZ9194
(18)
A similar flow chart can be adpated for the
conventional memory storage and recall of entered and
translated words in sequence. Irregular verbs and
plurals can be called at any time that input word re-
malns ln random access memory~
Referring to Figure 5, an alternate embodiment
208 is shown to illustrate the modifications that may be
necessary to accomodate presently available components
or the necessities of certain language problems. In
Figure 5, is shown a hand-held processor 209 and keyboard
210 with a dual display 212. The processor includes the
basic processors and components such as the CPU, the key-
board display interface, the clock generator, the system
controller, the random access memory, the peripheral
interface and the speech controller. The keyboard includes
on-off switches 214 and 216 for power supply and speech
circuits respectively, as well as function keys 217 simi-
lar to those in the previous embodiment.
The keyboard and processor are connected by an
electrical cable 218 to a pack 220. The pack is divided
into three sections or compartments. A top compartment
222 includes the audio filter, amplifier, and accompanying
circuits of the audio section of speech synthesizer and a
speaker 224 for sound reproduotion. The read only memory
is located in a casette in a central section 226. This
casette can be designed in a module fashion such that it
is replaceable by other similar casettes for different
language matches. At the base of the pack 226 is a power
supply 228 with a recharge terminal 230. The power supply
228 for both the preferred translator device 10 and the
: .
, (19)
alternate embodiment 208 is a rechargeable nickel-cadmium
battery which is preferred for its long life and recharging
capabilities using a conventional d.c. transformer (not
shown) and a.c. power source.
A clip 225 permits the pack 220 to be convenient-
ly carried in a pocket. Terminal 227 allows auxiliary
output units such as a printer or cathode ray tube (not
shown) to be connected to the unit if desired.
The pack 220 is primarily sized to accommodate
the necessary memory packet. Using bubble memory rather
than present shelf I.C. technology for memory circuits
the space requirements can be greatly reduced. It is
contemplated that the processors for the display transla-
tions and the speech synthesis can be reduced to a limited
number of custom I.C. chips.
In the alternate embodiment, the keyboard 234
is greatly expanded from that shown for the preferred
embodiment. This expansion is required for those instances
where one or both of the languages involved for transla-
tion has a character requirement in excess of the standard
26 for the English and Romance Language Countries.
While in the foregoing specification embodiments
of the invention have been set forth in considerable detail
for purposes of making a complete disclosure of the inven-
tion, it will be apparent to those skilled in the art that
numerous changes may be made in such details without de-
parting from the spirit and principles of the invention.
~'
