Note: Descriptions are shown in the official language in which they were submitted.
CA 02496872 2005-02-10
PHONETIC AND STROKE INPUT METHODS OF CHINESE
CHARACTERS AND PHRASES
BACKGROUND OF THE INVENTION
TECHNICAL FIELD
This invention relates generally to text entry technology. More particularly,
the invention
relates to a system and method for inputting Chinese characters and phrases.
DE;3CRIPTION OF THE PRIOR ART
For many years, the keyboard size has been a major size-limiting factor in the
efforts to
design and manufacture small portable computers because if standard typewriter-
size
keys are used, a portable computer must be at least as large as the keyboard.
Although
many kinds of miniaturized kE:yboards have been used in portable computers,
they have
been found too small to be easily or quickly manipulated by a regular user.
Incorporating a full-size keyboard in a portable computer also hinders true
portable use
of the computer. Most portable computers cannot be operated without placing
the
computer on a substantially flat work surface to allow the user to type with
both hands.
The user cannot easily use a portable computer while standing or moving. In
the latest
generation of small portable computers, called Personal Digital Assistants
(PDAs) or
palm-sized computers, manufacturers have attempted to address the problem by
incorporating handwriting recognition software in the device. Users may
directly enter
text by writing on a touch-sensitive panel or screen. The handwritten text is
then
converted by the recognition software into digital data. Unfortunately, in
addition to the
fact that printing or writing with a pen is usually slower than typing, the
accuracy and
speed of the handwriting recognition software has to date been less than
satisfactory. In
the case of Chinese language, with its large number of complex characters, the
issue
becomes especially complex. To make matters worse, today's handheld computing
1
CA 02496872 2005-02-10
devices which require text input are becoming smaller still. Recent advances
in two-way
paging, cellular telephones, and other portable wireless technologies have led
to a
demand for small and portable two-way messaging systems, and especially for
systems
which can both send and receive electronic mail ("e-mail")
Pinyin input method is one of the most commonly used Chinese character input
method
based on Pinyin, the official system of sounds forming syllables for Chinese
language
which was introduced in 1958 by the People's Republic of China. It is
supplementary to
the 5,000-year-old traditional Chinese writing system. Pinyin is used in many
different
ways. For examples: it is used as a pronunciation tool for language learners;
it is used in
index systems; and it is used for inputting Chinese characters into a
computer. The
Pinyin system adopts the standard Latin alphabets and takes the traditional
Chinese
analysis of the Chinese syllatale into initials, finals (ending sounds) and
tones.
Mandarin Chinese has consonant sounds that are found in most of the languages.
For
example, b, p, m, f, d, t, n, I, c,~, k, h are quite close to English. Other
initial sounds, such
as retroflex sounds zh, ch, sh and r, palatal sounds j, q and x, as well as
dental sounds z,
c and s, are different from English or Latin pronunciation. Table 1 lists all
initial sounds
according to the Pinyin system.
Table 1. Initial Sounds
Initial SoundPronunciation sample Note
I
Group I: Same
pronunciation
.as in English
M Man
N No
L Letter
F From
S Sun
W Woman
Y Yes
2
CA 02496872 2005-02-10
Group II:
Slightly
Different
from English
Pronunciation
P Pun use a strong puff of breath
K Cola use a strong puff of breath
T Tongue use a strong puff of breath
B Bum no puff of breath
D Dung no puff of breath
G Good no puff of breath
H Hot slightly more aspirated
than in
English
Group III:
Different
from English
Pronunciation
ZH Jeweler
CH As in ZH but with a strong
puff of
breath
SH Shoe
R Run
C Like "ts" in "it's high",
but with a
strong puff of breath
J Jeff
Q Close to "ch" in "Cheese"
X Close to "sh" in "sheep"
The finals connect with the initial sounds to create a Pinyin syllable which
corresponds to
a Chinese character (zi: ~). A Chinese phrase (ci: i~) usually consists of two
or more
3
CA 02496872 2005-02-10
Chinese characters. Table 2 lists all the final sounds according to the Pinyin
system and
Table 3 gives some examples illustrating the combination of initials and
finals.
Table 2. Final (ending) Sounds
Final Sound Pronunciation sample
a As in father
an Like the sounds of "Anne"
ang Like the sound "an" with addition of "g"
ai As in "high"
ao As in "how"
ar As in "bar"
o Like "aw"
ou Like the "ow" in "low"
ong Like the ''ung" in "jungle" with a slight
"oo" sound
a Sounds like "uh"
en Like the "un" in "under"
eng Like the "ung" in "lung"
ei Like the "ei" in "eight"
er Like the "er" in "herd"
i Like the "i" in machine
in As in "bin"
ing Like "sing"
a Like the "oo" in "loop"
un As in "fun"
Table 3. Putting Initials and Final {ending) Together
Pinyin Pronunciation sample
Ni Like "knee"
4
CA 02496872 2005-02-10
Hao Like "how" with a little more aspiration
Dong Luke "doong"
Qi Like "Chee"
Gong Like "Gung"
Tai Like "Tie"
Ji Like "Gee"
Quan Like "Chwan"
Each Pinyin pronunciation has one of the five tones (four pitched tones and a
"toneless"
tone) of Mandarin Chinese. A tone is important to the meaning of the word. The
reason
for having these tones is probably that Chinese language has very few possible
syllables
-- approximately 400 -- while English has about 12,000. For this reason, there
may be
more homophonic words, i.e. words with the same sound expressing different
meanings,
in Chinese than in most other languages. Apparently tones help the relatively
small
number of syllables to multiply and thereby alleviate but not completely solve
the
problem. There is no paralleling concept of the tones in English. In English,
an incorrect
inflection of a sentence can render the sentence difficult to understand. But
in Chinese
an incorrect intonation of a single word can completely change its meaning.
For
example, the syllable "da" may represents several characters such as #~ in
first tone
(dal ) meaning "to hang over something", o in second tone (da2) meaning "to
answer",
~7 in third tone (da3) meaning "to hit", and ~c in fourth tone (da4) meaning
"big". The
numbers after each of the syllables indicates the tones. The tones are also
indicated by
marks such as da da da da. Table 4 shows a description of five tones for the
syllable
"da".
Table 4. Five Tones
Tone Mark Description
1S da High and level
2" da Starts mE:dium in tone, then rises to the
top
5
CA 02496872 2005-02-10
3' da Starts low, dips to the bottom, then rises
toward the
top
4 da Starts at the top, then falls sharp and
strong to the
bottom
Neutral da Flat, with no emphasis
To enter a Chinese character using the Pinyin system, the user selects English
letters
corresponding to the character's Pinyin spelling. For example, on a standard
QWERTY
keyboard, when the user wants a Chinese character with a Pinyin of "ni", he
needs to
press the "N" key and then the "I" key. After the "N" key and the "I" key are
pressed, a
list of Chinese characters associated with the Pinyin spelling "NI" is
displayed. Then, the
user selects the intended character from the list. This method is hereby
referred as the
basic Pinyin input method.
Five-stroke input method is another most commonly used method for inputting
Chinese
characters. Five-stroke is a shape-based input method which is based on the
structure,
or shape, of characters rathE:r than on their pronunciation. The main concept
behind
five-stroke input method is that characters can be built by combining roots.
Five-stroke
method allots some 200 radicals, or roots, to five sections corresponding to
five types of
character strokes in the Chinese writing system: lateral, vertical, left
sweep, dot/right
sweep and bend.
In other words, the five-stroke input method divides the set of roots and the
keyboard into
five main categories according to the shape of the first stroke used to write
each
character. Each of the five roots is further divided into five levels. The
resulting 25 root
categories are assigned to the 25 keys A-Y on the keyboard.
The user needs no more than four keystrokes to enter any character in the code
chart,
and the most frequently used G00 characters require only one or two
keystrokes. The
6
CA 02496872 2005-02-10
user must know which radicals are assigned to each key, but once the array is
memorized, the user can type quickly and accurately.
Since both the Pinyin input method and the five-strike input method are widely-
used
input methods for inputting Chinese characters and phrases, it is a common
marketing
requirement for a system to support both input methods. However, due to the
difference
of natural of phonetic-based input method and stroke-based input method, a
different set
of data will be required for each input method. The size of data is usually
very large and
at times it is usually difficult to support more than one set of data which
are input method
specific. This is especially true on capacity-limited devices such as reduced
keyboard
systems.
An effective reduced keyboard input system for Chinese language must satisfy
all of the
following criteria. First, the input method must be easy for a native speaker
to
understand and learn to use. Second, the system must tend to minimize the
number of
keystrokes required to enter' text in order to enhance the efficiency of the
reduced
keyboard system. Third, thE: system must reduce the cognitive load on the user
by
reducing the amount of attention and decision-making required during the input
process.
Fourth, the approach should minimize the amount of memory and processing
resources
needed to implement a practical system.
In addition, the system should support both phonetic-based and stroke-based
input
methods on a reduced keyboard system. The system should share phonetic and
stroke
data to minimize the increa;>e of data size so that the system only requires a
little
increase in storage capacity.
The basic Pinyin method can be applied to a reduced keyboard input system when
combined with a non-ambiguous method of input Latin alphabets such as the
multi-tap
method. All non-ambiguous method, however, requires lots of key strokes, which
is
burdensome when combined with the basic Pinyin method. Thus it is preferable
to
combine the basic Pinyin method with a disambiguating system. One approach is
developed to disambiguate only one Pinyin syllable at one time by requiring
the user to
7
CA 02496872 2005-02-10
select a delimiter key, such as key 1 or key 0, between Pinyin spellings that
correspond
to multiple Chinese characters in commonly known Chinese phrases (i~]~~, i.e.
a word
with more than one character). The selection of the delimiter key instructs
the processor
to search for Pinyin syllables that match the input sequence and for Chinese
characters
associated with the first Pinyin syllable which may be selected by default. As
shown in
FIG. 1, the user is trying to input the Chinese characters associated with the
Pinyin
spellings NI and Y. To do thia, the user would first select the '6' key 16,
then the '4' key
14. In order to instruct the processor to perform a search for a syllable
matching the
keys entered, the user then selects the delimiter key 10 and finally the '9'
key 19.
Because this process requires a delimiter key depression between commonly
linked
multiple Chinese character words, time is wasted.
What is needed is a new technique for inputting Chinese using phonetic-based
or stroke-
based method in a reduced keyboard.
SIUMMARY OF THE INVENTION
A system and method for inputting Chinese characters using phonetic-based or
stroke-
based input method in a reduced keyboard is disclosed. By introducing common
indices
to ideographic characters, the system allows the ideographic characters to be
shared
among different type of input methods such as phonetic-based input method and
stroke-
based input method. The system matches input sequences to input method
specific
indices such as phonetic or stroke indices. These input method specific
indices are then
converted into indices to ideographic characters, which is then used to
retrieve
ideographic characters.
In one preferred embodiment, a method for input ideographic characters with a
user
input device is disclosed. The user input device includes: (1 ) a plurality of
input means,
each of which being associated with a plurality of strokes or phonetic
characters, an input
sequence being generated each time when an input is selected by the user input
device;
8
CA 02496872 2005-02-10
(2) data consisting of a plurality of input sequences and, associated with
each input
sequence, an input method apecific database containing a plurality of input
sequences
and, associated with each input sequence, a set of phonetic sequences whose
spellings
correspond to the input sequence or a set of strokes sequences corresponding
to the
input sequence; and (3) an ideographic database containing a set of
ideographic
character sequences, wherein each ideographic character contains an
ideographic
index, a plurality of stroke indices to corresponding stroke sequences and a
plurality of
phonetic indices to corresponding phonetic sequences.
The method includes the steps of: entering an input sequence into a user input
device;
comparing the input sequence with the input method specific database and
finding
indices to matching strokes entries or phonetic entries and the matching
stroke entries or
phonetic entries; converting ~rhe matching indices to stroke entries or
phonetic entries to
matching ideographic indices; retrieving matching ideographic character
sequences
from the ideographic database by the matching ideographic indices; and
optionally
displaying one or more of the matched ideographic character sequences.
In another preferred embodirnent, a system is disclosed for receiving input
sequences
entered by a user and generating textual output in Chinese language. The
system
includes: (1 ) a user input device having a plurality of input means, each of
which being
associated with a plurality of strokes or phonetic characters, an input
sequence being
generated each time when an input is selected by the user input device; (2) an
input
method specific database containing a plurality of input sequences and,
associated with
each input sequence, a set of phonetic sequences whose spellings correspond to
the
input sequence or a set of strokes sequences corresponding to the input
sequence; (3)
an ideographic database containing a set of ideographic character sequences,
wherein
each ideographic character contains an ideographic index, a plurality of
stroke indices to
corresponding stroke sequences and a plurality of phonetic indices to
corresponding
phonetic sequences; (4) means for comparing the input sequence with the input
method
specific database and finding indices to matching strokes entries or phonetic
entries and
the matching stroke entries or phonetic entries; (5) means for converting the
matching
9
CA 02496872 2005-02-10
indices to stroke entries or phonetic entries to matching ideographic indices;
(6) means
for retrieving matching ideographic character sequences from the ideographic
database
by the matching ideographic indices; and (7) an output device for displaying
one or more
matched stroke or phonetic entries, and matched ideographic characters.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is schematic diagram showing a keyboard layout for inputting Chinese
characters
using delimiters between Pinyin syllables according to prior art;
F1G. 2 is a schematic view of an exemplary embodiment of a cellular telephone
which
incorporates a phonetic input method to a reduced keyboard system according to
the
invention;
FIG. 3 is schematic diagram depicting an exemplary display where tones are
used with
Pinyin spelling during inputting Chinese phrases;
FIG. 4 is a block diagram illu strating the hardware components of the reduced
keyboard
system of FIG. 2;
FIG. 5 is a block diagram illustrating a system for supporting both phonetic-
based and
stroke-based input method for generating textual output in Chinese language
according
to one preferred embodiment of the invention;
FIG. 6 is a block diagram illustrating an ideographic language text input
system
incorporated in a user input device according to one preferred embodiment of
the
invention;
FIG. 7 is a flow diagram illustrating a method for generating textual output
in Chinese
language using the system in FIG. 5; and
CA 02496872 2005-02-10
FIG. 8 is a flow diagram illustrating a phonetic input method for generating
textual output
in Chinese language according to one preferred embodiment of the invention.
DETAILED DESCRIPTION OF THE INVENTION
First referring to FIG. 5, which illustrates a system for supporting both
phonetic-based
and stroke-based input method is depicted for receiving input sequences
entered by a
user and generating textual output in Chinese language according to one
preferred
embodiment of the invention. The system includes the following:
~ a user input device 510 having a plurality of input means, wherein an input
sequence is generated each time when an input is selected by the user input
l0 device;
~ a database 520 containing a plurality of input sequences and, associated
with
each input sequence, a set of phonetic sequences whose spellings correspond to
the input sequence or a set of strokes sequences corresponding to the input
sequence;
Note that the stroke indices are typically indices of strokes sorted by stroke
sequences in
a stroke input system. The stroke input system can be a five-stroke or an
eight-stroke
system. The phonetic indices can be typically indices of phonetic characters
sorted by
actual spelling in a phonetic input system. The phonetic input system can be a
Pinyin
system or a Zhuyin system. Alternatively, the phonetic indices can be indices
of input
means in a phonetic input system.
~ a database 530 containing a set of ideographic character sequences, wherein
each ideographic character contains an ideographic index, a plurality of
stroke
indices to corresponding stroke sequences and a plurality of phonetic indices
to
corresponding phonetic; sequences;
11
CA 02496872 2005-02-10
Note that by introducing the indices to ideographic characters, the system
allows the
ideographic characters to be shared among different type of input methods such
as
phonetic-based input method and stroke-based input method. The database 530
also
contains information that is needed to convert between indices to ideographic
characters
and stroke indices, between indices to ideographic characters and phonetic
indices, and
from indices to ideographic characters to ideographic characters. These
ideographic
characters can be Unicode of GB code.
~ means for comparing the input sequence with the input method specific
database
and finding indices to matching strokes entries or phonetic entries and the
matching stroke entries or phonetic entries 540;
~ means for converting the matching indices to stroke entries or phonetic
entries to
matching ideographic indices 550;
~ means for retrieving matching ideographic character sequences from the
ideographic database by the matching ideographic indices 560; and
~ an output device 570 for displaying one or more matched phonetic entries and
matched ideographic characters.
FIG. 7 illustrates a method for generating textual output in Chinese language
using the
system in FIG. 5 according to one preferred embodiment of the invention. The
method
includes the steps of:
Step 710: Enter an input sequence into user input device 510;
In this step, a user first generates an input sequence using the input means
of the input
device 510.
Step 720: Compare the input sequence with input method specific database 520
and
find indices to matching strokes entries or phonetic entries and the matching
stroke
entries or phonetic entries;
12
CA 02496872 2005-02-10
In this step, based on the input method selected, the system uses the
comparing and
matching means 540 to find one or more indices to phonetic entries from the
database
520, or one or more indices to stroke entries.
Step 730: Convert the matching indices to stroke entries or phonetic entries
to
matching ideographic indices;
In this step, the system uses the converting means 550 to convert the matched
phonetic
entries or stroke entries to indices to matching ideographic characters.
Step 740: Retrieve matching ideographic character sequences from the
ideographic
database by the matching ideographic indices; and
In this step, the indices to matching ideographic characters are passed to the
retrieving
means 560 to retrieve matching ideographic characters.
Step 750: Optionally display one or more of the matched ideographic character
sequences.
In this step, the matched ideographic characters may be displayed on the
output device
570. One of the matched ide~ographic characters, such as the one with highest
FUBLM
value, is selected by default. The user may accept the default or select a
different
matched ideographic sequence.
FIG. 6 illustrates an ideographic language text input system incorporated in a
user input
device according to one prefen~red embodiment of the invention. The system
includes the
following:
~ a plurality of inputs 610, each of which associated with a plurality of
characters, an
input sequence being generated each time when an input is selected by
manipulating the user input device 605, wherein a generated input sequence
corresponds to a sequence of inputs that have been selected;
13
CA 02496872 2005-02-10
~ at least one selection input 620 for generating an object output, wherein an
input
sequence is terminated when the user manipulates the user input device to a
selection input;
~ a memory 630 containing a plurality of objects, wherein each of the
plurality of
objects is associated with an input sequence;
~ a display 640 to depict system output to the user; and
~ a processor 650 coupled to the user input device 605, memory 630, and
display
640.
The processor 650 further includes: identifying means 652 for identifying from
the
plurality of objects in the memory any object associated with each generated
input
sequence; output means 654 for displaying on the display the character
interpretation of
any identified objects associated with each generated input sequence; and
selection
means 656 for selecting the desired character for entry into a text entry
display location
upon detecting the manipulation of the user input device to a selection input.
Once the user manipulates the user input device 605 and selects the inputs
610, an input
sequence is generated. The processor 650 uses the identifying means 652 to
match one
or more linguistic objects from memory 630 with the generated input sequence.
The
character interpretation of the matched objects is output to the display 640
by the
processor 650 using the output means 654. The user then selects a character
interpretation with the selection input 620 and the processor 650 invokes the
selection
means 656 to output the selecaed character to a text entry display location.
Now referring to FIG. 2, which is a schematic view of an exemplary embodiment
of a
cellular telephone that incorporates a phonetic input method to a reduced
keyboard
system according to the invention. The portable cellular telephone 52 has a
display 53
and contains a reduced keyboard 54 implemented on the standard telephone keys.
For
14
CA 02496872 2005-02-10
the purposes of this invention, the term "keyboard" is defined broadly to
include any input
device including a touch screen having defined areas for keys, discrete
mechanical keys,
membrane keys, and the like. The arrangement of the Latin alphabets on each
key in the
keyboard 54 is corresponding to what has become a de facto standard for
American
telephones. Note that keyboard 54 thus has a reduced number of data entry keys
as
compared to a standard QWERTY keyboard, where one key is assigned for each
Latin
alphabet. More specifically, the preferred keyboard shown in this embodiment
contains
ten data keys numbered '1' through '0' arranged in a 3-by-4 array, together
with four
navigation keys comprising of Left Arrow 61 and Right Arrow 62, Up Arrow 63
and Down
Arrow 64.
The user enters data via keystrokes on the reduced keyboard 54. In the first
preferred
embodiment, when the user enters a keystroke sequence using the keyboard, text
is
displayed on the telephone display 53. Three regions are defined on the
display 53 to
display information to the user. A text region 71 displays the text entered by
the user,
serving as a buffer for text input and editing. A phonetic, e.g. Pinyin,
spelling selection
list 72, typically located below the text region 71, shows a list of Pinyin
interpretations
corresponding to the keystrolke sequence entered by the user. A phrase
selection list
region 73, e.g. Chinese phrases, typically located below the spelling
selection list 72,
shows a list of words corresponding to the selected Pinyin spelling, which is
corresponding to the sequence entered by the user. The Pinyin selection list
region 72
aids the user in resolving the ambiguity in the entered keystrokes by
simultaneously
showing both the most frequently occurring Pinyin interpretation of the input
keystroke
sequence and other less frequently occurring alternate Pinyin interpretations
displayed in
descending order of FUBLM. The Chinese phrase selection list region 73 aids
the user
in resolving the ambiguity in the selected Pinyin spelling by simultaneously
showing both
the most frequently occurring Phrase text of the selected spelling and other
less
frequently occurring Phrase text displayed in descending order of frequency of
user base
on a linguistic model (FUBLM). While Pinyin is described herein as comprising
a
phonetic input, it should be appreciated that phonetic inputs may comprise
Latin
alphabet; Bopomofo alphabet also known as Zhuyin; digits; and punctuation.
CA 02496872 2005-02-10
In order to present the user with possible phrases, the system relies on a
linguistic model
which can be limited to words found exactly in a database ordered
alphabetically or
according to total number of keystroke in ideographs, radicals of ideographs
or a
combination of both. The linguistic model can be extended to order linguistic
objects
according to a certain fixE:d frequency of common usage such as in formal or
conversational, written or conversational spoken text. Additionally, the
linguistic model
can be extended to use N-gram data to order particular characters. The
linguistic model
can even be extended to use grammatical information and transition frequencies
between grammatical entities to generate phrases which go beyond those phrases
included in the database. Thus the linguistic model may be as simple as a
fixed
frequency of use and a fixed number of phrases, or include adaptive frequency
of use,
adaptive words or even involve grammatical/semantic models which can generate
phrases that go beyond those contained in the database.
Referring to FIG. 4, which schematically depicts the hardware components of
the
reduced keyboard system of I=IG. 2, the keyboard 54 and the display 53 are
coupled to a
processor 100 through appropriate interfacing circuitry. Optionally, a speaker
102 is also
coupled to the processor 100. The processor 100 receives input from the
keyboard 54,
and manages all output to thE: display 53 and speaker 102. Processor 100 is
coupled to
a memory 104. The memory 104 includes a combination of a temporary storage
media,
such as random access memory (RAM), and a permanent storage media, such as
read-
only memory (ROM), floppy disks, hard disks, or CD-ROMs. Memory 104 contains
all
software routines to govern system operation. Preferably, the memory 104
contains an
operating system 106, disambiguating software 108, and associated vocabulary
modules
110 which are discussed above. Optionally, the memory 104 may contain one or
more
application programs 112, 1'14. Examples of the application programs include
word
processors, software dictionaries, and foreign language translators. Speech
synthesis
software may also be provided as an application program which allows the
reduced
keyboard disambiguating system to function as a communication aid.
16
CA 02496872 2005-02-10
Referring back to FIG. 2, the reduced keyboard system allows a user to quickly
enter text
or other data using only a single hand. The user enters data using the reduced
keyboard
54. Each of the data keys 2 through 9 has multiple meanings, represented on
the top of
the key by Latin alphabets, numbers, and other symbols. Because individual
keys have
multiple meanings, keystroke sequences are ambiguous as to their meaning. When
the
user enters data, the various keystroke interpretations are therefore
displayed in multiple
regions on the display 53 to aid the user in resolving any ambiguity. On large-
screen
devices, a Pinyin selection list of possible interpretations of the entered
keystrokes and a
Chinese phrase selection list of the selected Pinyin spelling are displayed to
the user in
the selection list regions. T'he first entry in the Pinyin selection list is
selected as a
default interpretation and highlighted in any way to distinguish itself from
the other Pinyin
entries in the selection list. In the preferred embodiment, the selection
Pinyin entry is
displayed in reverse color image such as white font with a dark background.
The Pinyin selection list of the possible interpretations of the entered
keystrokes may be
ordered in a number of ways. In a normal mode of operation, the keystrokes are
initially
interpreted as a Pinyin spelling consisting of complete Pinyin syllables
corresponding to
a desired Chinese phrase (hereinafter as complete Pinyin interpretation). As
keys are
entered, a vocabulary module look-up is simultaneously performed to locate
valid Pinyin
spellings corresponding to the input key sequence. The Pinyin spellings are
returned
from the vocabulary module according to FUBLM, with the most commonly used
Pinyin
spelling listed first and selected by default. The Chinese phrases matching
the selected
Pinyin spelling are also returned from the vocabulary module according to
FUBLM.
Normally the user can find the Chinese phrase he wants to input in the Chinese
phrase
select list and then select the Chinese phrase and input the Chinese phrase in
the text
input region 71. If the default selected Pinyin spelling is what the user
wants to input, but
the Chinese phrase he wants to input is not displayed, he can use the Up Arrow
63 and
Down Arrow 64 keys to display an extended set of other matched Chinese phrases
from
the vocabulary database. In a few cases, the Pinyin selection list region 72
cannot hold
all matched Pinyin spellings, and thus the Left Arrow 61 and Right Arrow 62
keys are
used to scroll the previously off-screen Pinyin spellings into the Pinyin
select list region
17
CA 02496872 2005-02-10
72. For example, if the default selected Pinyin spelling is not what the user
wants to
input, he can use the Left Arrow 63 and Right Arrow 64 keys to select other
matched
Pinyin spellings.
In the majority of text entry, keystroke sequences are intended by the user to
spell out
complete Pinyin syllables. It is appreciated, however, that the multiple
characters
associated with each key alilow the individual keystrokes and keystroke
sequences to
have several interpretations. In the preferred reduced keyboard disambiguating
system,
various different interpretations are automatically determined and displayed
to the user
as a list of Pinyin spellings and a list of Chinese phrases corresponding to
the selected
Pinyin spellings.
For example, the keystroke sequence is interpreted in terms of partial Pinyin
spelling
corresponding to possible Chinese phrases that the user may be entering
(thereinafter
as partial Pinyin interpretation). Unlike complete Pinyin interpretation,
partial Pinyin
spelling allows the last Pinyin syllable to be incomplete. A Chinese phrase is
returned
from the vocabulary database if its Pinyin for the characters before the last
character
matches all syllables before the last partial Pinyin syllable while the Pinyin
syllable of the
last character starts with the ,partially completed syllable. By returning
Chinese phrases
that match a Pinyin spelling that extends the original partial phrasal Pinyin
with a
possible completion of the last Pinyin syllable, the partial Pinyin
interpretation allows the
user to easily confirm that the correct keystrokes have been entered, or to
resume typing
when his attention has been diverted in the middle of the phrase. The partial
Pinyin
interpretation is therefore provided as entries in the Pinyin spelling list.
Preferably, the
partial Pinyin interpretations are sorted according to the composite FUBLM of
the set of
all possible Chinese phrases that can match a Pinyin spelling that extends the
partial
Pinyin input with a possible completion of the last Pinyin syllable. Partial
Pinyin
interpretations provide feedb<~ck to the user by confirming that the correct
keystrokes
have been entered to lead to the entry of the desired word.
18
CA 02496872 2005-02-10
To reduce the number of possible matches displayed, the user may also input a
syllable
delimiter after a completed Pinyin syllable. In one preferred embodiment, the
'0' key is
used as a syllable delimiter. If syllable delimiters are entered, only Pinyin
spellings
whose syllable ending matches the position of syllable delimiters are returned
and
displayed in the Pinyin selection list region 72.
In another preferred embodiment, the user may also input a tone after each
completed
Pinyin syllable. After each completed Pinyin syllable, the user presses a tone
key
followed a number which corresponding to the tone of the syllable. In this
preferred
embodiment, the '1' key is used as the tone key. If tones are entered, only
Pinyin
spellings having ~ Chinese phrases conversions that match the tones are
returned and
displayed in the Pinyin selection list region 72. The displayed Pinyin
spellings also
include the tones that have been entered. As shown in FIG. 3, the Pinyin
spelling
"Bei3Jing1" is displayed in the Pinyin spelling list region 72. If a Pinyin
spelling with
tones has been selected, only Chinese phrases that match both the Pinyin
spelling and
the corresponding tones are returned and displayed. The filtering may be
applied to
tones following a complete Pinyin syllable or a partial Pinyin spelling.
The partial Pinyin completion looks ahead until the last syllable is complete.
There are
maximum five nodes in the second section of the path because the longest
syllable is
"Chuang" or "Shuang" or Zhuang". Only in these three cases, the process looks
ahead
five more nodes.
For instance, if the key input is "2345", one of the valid spellings is
"BeiJ". The first
complete syllable is "Bei". The second is "J" that is not a complete syllable.
Thus, the
first section of the path for this case is to build the spelling "BeiJ". The
process will look
ahead in the vocabulary module tree to complete the last syllable. Then, it
finds the word
(BeiJing) that has partial spelling matches "BeiJ". The second section of the
path is used
to build "ing". If the word "BeiJingShi" is also in the vocabulary module
tree, the process
would not locate this word for the key input "2345" because it requires
looking ahead two
more syllables.
19
CA 02496872 2005-02-10
If any tone is entered, the process can filter the characters because the
character tones
are retrieved along with their Unicodes when secondary instructions are
executed. If a
character has more than one pronunciation, the most common one is retrieved
first.
The conversions (characters and words) for each spelling are prioritized by
the FUBLM.
The most frequently used character or word is retrieved first during the
spelling-
character/word conversion. The words converted from the exactly matched
spelling are
ordered ahead of the words converted from the partial matched spellings. The
words
converted from the different partial matched spellings are sorted by the key
order (that is,
key 2, 3, 4, 5...) and the frequency order of the letters on the key
(character on the key
index). For example, assuming the active spelling is "Sha", because 'n' is
ordered ahead
of 'o' when the previous letter is 'a', the characters converted from the
"Sha" are returned
first, followed by these converted from "Shai", "Shan", "Shang" and "Shao".
FIG. 8 illustrates a phonetic; input method for generating textual output in
Chinese
language according to one preferred embodiment of the invention. The method
includes
the steps of:
Step 810: Enter an input sequence into a user input device;
Step 820: Compare the input sequence with the phonetic sequence database and
find matching phonetic entries and their indices;
Step 830: Display optionally one or more matched phonetic entries;
Step 840: Convert "indices to phonetic entries" to "indices to ideographic
characters"
and retrieve matching ideographic characters from the ideographic database by
the
indices to ideographic characters; and
Step 850: Optionally display one or more matched ideographic characters.
CA 02496872 2005-02-10
In another preferred embodiment, the disambiguating Pinyin system allows
spelling
variations which are typically caused by regional accents. Regional accents
can lead to
variations in pronunciations 'for various syllables. This can lead to
confusion about for
instance "zh " and z-", "-n" and "-ng." To accommodate these variations,
variations on
certain spellings can be considered. Variations can either be displayed as
part of the
selection list for the particular Pinyin, for instance if the user types "zan"
the selection list
may include "zhan" and "zhang" as possible variants, or the user when failing
to find a
particular character may select a "show variants" options which will provide
the user with
possible variations of the spE;lling. Additionally the user may be able to
turn off and on
particular "confusion sets" such as "z <-> zh", "an <-> ang" etc.
Table 5. Examples of Common Confusion Sets
A la
E IE
O Ou, uo
An Ang, ian, fang
En Eng
-
In Ing
Ong long
Uan Uang
On Ong, iong
Ao lao
Z Zh
C Ch
S _ S h
L N
21
CA 02496872 2005-02-10
In another preferred embodiment, the disambiguating system includes a custom
word
dictionary. Since the dictionary of phrases is limited by the available
memory, the
custom word dictionary is essential that the user can add Pinyin/character
combinations
manually which can then be accessed via the input method.
In another preferred embodiment, the disambiguating Pinyin system may update
the
FUBLM adaptively based on the recency of use. The initial phrases are ordered
according to a particular linguistic model (for instance the frequency of use
in a corpus)
which may not match the user's expectations. By tracking the user's patterns,
the
system will learn and update the linguistic model accordingly.
In another preferred embodiment, the system may provide the user with word
predictions
based on the words syllables entered so far and a linguistic model. The
linguistic model
may be used to determine in which order the predictions should be presented to
the
user. In fact the linguistic model can provide the user with predictions of
words even
before the user types any characters. Such a linguistic model may be based on
simple
frequency of use of single characters, or frequency of use of two or more
character
combinations (N-grams) or a grammatical model or even a semantic model. In
alternative embodiments; the number of total keystrokes in an ideograph;
radical of an
ideograph; radical and number of strokes of a radical; alphabetically ordered;
frequency of occurrence of ideograph sequences or phonetic sequences in
formal,
conversational written, or conversational spoken text; frequency of occurrence
of
ideographic sequences or phonetic sequences when following a preceding
character or
characters; proper or common grammar of the surrounding sentence; application
context of current input sequence entry; and recency of use or repeated use of
phonetic
or ideographic sequences by the user or within an application program.
While the preferred input method would require the user to enter the full
spelling of the
word, the user may select to enter only the first character of each syllable.
Thus instead
of typing BeiJing, the user type BJ and is provided with phrases that match
this acronym.
Additionally, the user may define their own acronyms and add them to the
Custom word
dictionary.
22
CA 02496872 2005-02-10
In addition to ambiguous entry of characters, the system may also provide a
non-
ambiguous method for the user to explicitly select a character.
During the input process, the user may enter partial syllables for each of the
multiple
syllable words. Preferably, the number of partial keystrokes for each syllable
is one, for
example, the first keystroke of each syllable.
The system may also display the valid final sounds after the user identifies
the initial
sound. For example, if a user is trying to input Pinyin syllable "Zhang", the
user first
identifies the initial sound "zh" and then is provided with valid final sounds
for the initial
for which the user may select "ang".
During the input process, thE: user may also select one of the many inputs
associated
with a special wildcard input. The special wildcard input may match zero or
one of
phonetic characters.
The system may also display phonetic sequences that include matching entries
in
English or other alphabetic languages and allow simultaneous interpretation of
the key
presses as syllables and words in a secondary language such as English.
As is shown by the above detailed description, a system has been designed to
create an
effective reduced keyboard input system for Chinese language. First, the
method is easy
for a native speaker to understand and learn how to use because it is based on
the
official Pinyin system. Second, the system tends to minimize the number of
keystrokes
required to enter text. Third, the system reduces the cognitive load on the
user by
reducing the amount of attention and decision-making required during the input
process
and by the provision of appropriate feedback. Fourth, the approach disclosed
herein
tends to minimize the amount of memory and processing resources required to
implement a practical system.
23
CA 02496872 2005-02-10
Those skilled in the art will also recognize that minor changes can be made to
the design
of the keyboard arrangement and the underlying database design, without
significantly
departing from the underlying principles of the current invention.
Accordingly, the invention should only be limited by the Claims included
below.
24
