WO2013053114A1 - Translation Apparatus - Google Patents

Translation Apparatus Download PDF

Info

Publication number
WO2013053114A1
WO2013053114A1 PCT/CN2011/080710 CN2011080710W WO2013053114A1 WO 2013053114 A1 WO2013053114 A1 WO 2013053114A1 CN 2011080710 W CN2011080710 W CN 2011080710W WO 2013053114 A1 WO2013053114 A1 WO 2013053114A1
Authority
WO
WIPO (PCT)
Prior art keywords
batch
database
translation
batches
series
Prior art date
Application number
PCT/CN2011/080710
Other languages
French (fr)
Inventor
Kak San PANG
Original Assignee
Genius Link International Limited
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Genius Link International Limited filed Critical Genius Link International Limited
Priority to PCT/CN2011/080710 priority Critical patent/WO2013053114A1/en
Publication of WO2013053114A1 publication Critical patent/WO2013053114A1/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F40/00Handling natural language data
    • G06F40/40Processing or translation of natural language
    • G06F40/42Data-driven translation
    • G06F40/47Machine-assisted translation, e.g. using translation memory

Definitions

  • the present invention relates to translation apparatuses, and in particular, but not limited to, translation apparatuses for translating phrases used in everyday conversation.
  • the invention will be described in relation to portable personal translation devices, but is not limited to this particular use.
  • Existing translation devices include those that translate phrases used in everyday conversation from a source language to a target language. These types of devices are particularly suited for personal use in situations that require the real-time translation of phrases, such as when a user engages in conversation with a person who does not share a common language with the user, and during the ensuing conversation.
  • Existing translation devices typically include a database of phrases.
  • the databases are simply linear databases and users simply scroll through the phrases until the desired phrase is found. Once a desired phrase is found, the user must return to scrolling through the phrases to find the next desired phrase. This becomes very time-consuming and inefficient with a large number of phrases.
  • phrases are grouped under category or topic headings in the database. Users select the applicable topic heading and then scroll through the phrases under the selected topic heading until the desired phrase is found. However, once a desired phrase is found, the user must start from the original starting point again to find the next desired phrase. In particular, if the next desired phrase is not in the current topic heading, the user must again select a topic heading and then scroll though the phrases under the topic heading to find the next desired phrase. If phrases are grouped under levels of category or topic headings, then a user must drill down the levels to the desired group of phrases to find the desired phrase. If the next phrase is not in the current group, then the user must start from the same starting point again, that is, the top level of topic headings. Similar to the linear database example above, this is still very time-consuming and inefficient with a large number of phrases.
  • the present invention in a first aspect, provides a translation apparatus including a database of phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.
  • the database includes scenarios, each scenario linked to a respective batch such that selection of a scenario accesses a respective batch.
  • at least one scenario is a member of a batch.
  • the database includes one or more series of batches linked successively together whereby a primary member of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and one or more of the batches in a series has a secondary member that is linked to another batch other than the next batch in the series.
  • the database includes top-level scenarios, each top-level scenario linked to a respective series of batches such that selection of a top-level scenario accesses the first batch in a respective series of batches.
  • the translation apparatus includes a user input selected from one or more or any combination of the following: a keyboard, a switch, a computer mouse, a touchpad, a touch screen, an audio input device, a microphone, a scanning device, a barcode reader, a card reader, an optical input device, and an optical mark reader.
  • the translation apparatus includes an output selected from one or more or any combination of the following: a video output device, a monitor, a display screen, an audio output device, a speaker, an earphone jack, an earphone, and a printing device.
  • the translation apparatus includes a memory for storing the database, the memory selected from one or more or any combination of the following: a semiconductor storage device, a ROM IC, flash memory, a hard disk, and magnetic tape.
  • the translation apparatus is in the form of an integrated translation unit for use by a user, the integrated translation unit having an onboard memory and supplied with the database loaded on the onboard memory.
  • the translation apparatus includes a translation unit for use by a user and the database is stored on a remote memory.
  • the translation unit communicates with the remote memory to access the database.
  • the translation unit has an onboard memory and the translation unit downloads the database from the remote memory onto the onboard memory.
  • the translation unit performs the following steps: receive commands from a user; access the database according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and output the selected phrase translated into the at least one target language.
  • the translation unit has an onboard memory and the steps are in the form of an executable program stored on a remote memory, the translation unit downloading the executable program onto the onboard memory to execute the program to perform the steps.
  • the present invention provides a non-transitory computer-readable storage medium with a database stored thereon, the database containing phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.
  • the database includes scenarios, each scenario linked to a respective batch such that selection of a scenario accesses a respective batch.
  • at least one scenario is a member of a batch.
  • the database includes one or more series of batches linked successively together whereby a primary member of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and one or more of the batches in a series has a secondary member that is linked to another batch other than the next batch in the series.
  • the database includes top-level scenarios, each top-level scenario linked to a respective series of batches such that selection of a top-level scenario accesses the first batch in a respective series of batches.
  • the present invention provides a non-transitory computer-readable storage medium with an executable program stored thereon, wherein the program instructs a translation unit to perform the following steps: receive commands from a user; access the database described above according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and output the selected phrase translated into the at least one target language.
  • Fig. 1 is a block diagram of a translation apparatus in accordance with a preferred embodiment of the present invention
  • Fig. 2 is a schematic diagram of the network structure of phrases in the database of a translation apparatus in accordance with another preferred embodiment of the present invention
  • Fig. 3 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with another preferred embodiment of the present invention.
  • Fig. 4 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with yet another preferred embodiment of the present invention.
  • Fig. 5 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with yet another preferred embodiment of the present invention.
  • Fig. 6 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with a further preferred embodiment of the present invention.
  • Fig. 7 is a flow diagram of the operation of a translation apparatus in accordance with a preferred embodiment of the invention.
  • Fig. 8 is a table showing part of a database of a translation apparatus in accordance with a preferred embodiment of the invention.
  • Fig. 9 is a table showing another part of the database of Fig. 8.
  • the translation apparatus 1 includes a database 2 of phrases 3 for translation from a source language to at least one target language.
  • the phrases in the database 2 are members 4 of batches 5.
  • At least one member 4 in a parent batch is linked to a child batch such that the child batch is accessible by selecting the at least one member.
  • the parent and child batches can be any two of the batches.
  • the same batch 5 can be a parent batch and a child batch at the same time.
  • the term “parent” when referring to a batch as a “parent batch” simply means that a member 4 of the batch is linked to another batch, which is referred to as a “child” batch. This “child batch” can also be referred to as a “parent batch” if it too has a member 4 that is linked to another batch.
  • the database also includes scenarios 6, with each scenario linked to a respective batch 5 such that selection of a scenario accesses a respective batch. At least one scenario 6 is a member 4 of a batch 5.
  • a phrase 3 is a natural language expression typically used in conversation, such as “Where is the nearest drug store?” or “Are there any duty-free concessions?”, as shown in Fig. 6.
  • Phrases 3 are the members 4 that are primarily required to be translated by users.
  • a scenario 6 is a higher-level or general description of members of a batch, that is, the phrases or sub-scenarios in the batch.
  • a scenario is linked to the batch whose members are generally described by the scenario name so that selection of the scenario accesses the related batch.
  • a phrase or a scenario can be selected by a user either for translation or for accessing the batch linked to the scenario or, if any, the batch linked to the phrase.
  • scenario 6c “Drug Store Directions”.
  • Scenario “Drug Store Directions” 6c is itself a member of a batch 5c that includes other scenarios 6c “Opening Hours” and “Payment” as members.
  • This batch 5c is in turn linked to scenario 6b “Drug Store”, a member of a batch 5b which includes other scenarios “First Aid” and “Payment” as members.
  • scenario 6a “Medical & Emergencies”.
  • Scenarios 6c “Drug Store Directions”, “Opening Hours”, and “Payment” can be seen as sub-scenarios to scenario 6b “Drug Store”.
  • scenarios 6b “Drug Store”, “First Aid”, and “Payment” can be looked upon as sub-scenarios to scenario 6a “Medical & Emergencies”.
  • Scenario 6a “Medical & Emergencies” can be referred to as a “top-level” scenario.
  • a batch 5 can have a mix of phrases 3 and scenarios 6 as its members 4.
  • batch 5d includes the phrases 4d “How do I get there” and “Could you take me there” with the scenario 6d “Back to Drug Store”.
  • the batches can be described as going from a higher level batch 5b to progressively lower level batches 5c and 5d. As can be seen, higher level batches generally cover broader subject matter than lower level batches which cover more and more specific subject matter.
  • the links going from higher level batches to lower level batches can be referred to as forward links and the links going from lower level batches to higher level batches can be referred to as backward links.
  • Batches 5b and 5c are parent and child batches, respectively, since a member 4b of batch 5b, namely scenario 6b “Drug Store”, links to batch 5c.
  • Batch 5c is also a parent batch in relation to batch 5d which is a child batch in this instance, due to the linking of member 4c in the form of scenario 6c “Drug Store Directions” to batch 5d.
  • scenario 6c “Back to Drug Store”
  • batch 5d is also a parent batch and batch 5c is also a child batch.
  • the database 2 includes one or more series of batches 5x, 5y, 5z linked successively together whereby a primary member 4’ of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and wherein one or more of the batches in a series has a secondary member 4’’ that is linked to another batch other than the next batch in the series.
  • members 4 can also link to batches 5 in a different series.
  • links in the same series of batches can be forward links and backward links, members 4 can also link with batches 5 in a different series with forward links or backward links.
  • Fig. 2 there are also user-defined batches 5f. These batches contain phrases 3f or scenarios 6f from the database 2 marked as favourites by a user. These can be linked to user-defined scenarios 6e that have descriptions defined by the user.
  • the database further includes top-level scenarios 6TL, with each top-level scenario linked to a respective series of batches 5x, 5y, 5z such that selection of a top-level scenario accesses the first batch in a respective series of batches.
  • top-level scenarios 6TL “Medical & Emergencies” and “Transportation” link to first batches 5g and 5p, respectively.
  • the database structure described above is particularly useful for organizing phrases in logical flows according to natural everyday conversation.
  • the variety of linkages between phrases allows phrases that would follow each other in logical conversational flows to appear to the user in accordance with those logical flows or closely matching those logical flows.
  • the organization of phrases in batches and the linkages can be informed by human conversational habits. These can be undertaken by interviewing or surveying people in relevant situations, such as travelers and business people, and analyzing the results to determine typical conversational flows in a variety of situations.
  • a user at an airport looking for a pharmacy starts by selecting top-level scenario 6TL “Transportation” to access batch 5t with scenario members 6t “Air Travel” and “Travelling by Land”. The user then selects “Air Travel” to access batch 5u with scenario members 6u “Airport Shops Directions” and “Shopping”. The user then selects “Airport Shops Directions” to access batch 5v with phrase members 3v “Are there any duty-free concessions?” and “Is there a pharmacy?”. The user then selects the desired phrase “Is there a pharmacy?” for translation from the user’s source language to the target language of an airport staff member or a passer-by.
  • All of the phrases 3 in the database 2 are in multiple languages. All of the versions of the phrases 3 in the different languages are linked together so that when a phrase 3 in a source language is selected for translation to a target language, the version of the phrase 3 in the target language is provided for output.
  • batch 5k with scenario members 6k “Drug Store Directions”, “Opening Hours”, and “Payment” is accessed, which are scenarios that link to likely phrases that would be required in the next stage of the conversation, such as phrase member 3l “Where is the nearest drug store?” in batch 5l, which can be accessed by selecting scenario member 6k “Drug Store Directions” in batch 5k.
  • FIG. 8 An example of a specific implementation of the database 2 is shown in Fig. 8, which is in the form of a table depicting part of the database 2.
  • the first column lists the location IDs of members 4, each of which is in a parent batch 5.
  • the second column lists the location IDs of the members of the child batch 5 linked to a member 4 in a parent batch 5, with the member of the parent batch listed in the first column on the same row of the table.
  • the member 4 with the location ID “1010010002” in the first column is a member of a parent batch 5, and is linked to a child batch with members having location IDs “1010020002”, “1010020356”, “1010020666”, and “1010010002”.
  • the member 4 with the location ID “1010020002” in the first column is a member of a parent batch 5, and is linked to a child batch with members having location IDs “1010030002”, “1010030050”, “1010030187”, “1010030202”, “1010030261”, “1010030276”, “1010030283”, “1010030293”, and “1010010002”.
  • the third column lists the ID of the phrase or scenario having the location listed in the first column on the same row of the table.
  • the member 4 with the location ID “1010010002” in the first column has the phrase or scenario ID “0001”.
  • the fourth column lists the ID of the phrase or scenario having the location listed in the second column on the same row of the table.
  • the member 4 with the location ID “1010020666” in the second column has the phrase or scenario ID “9652”.
  • the location IDs are coded so that the first three digits of the location ID identify the top-level scenario 6TL, the second three digits identify a column in a part of the database, and the last four digits identify a row in a part of the database.
  • Fig. 9 shows the part of the database where the actual phrase or scenario wording in multiple languages can be found for the phrases or scenarios with IDs “0001” and “0002”.
  • the translation apparatus includes a user input 7 selected from one or more or any combination of the following: a keyboard, a switch, a computer mouse, a touchpad, a touch screen, an audio input device, a microphone, a scanning device, a barcode reader, a card reader, an optical input device, and an optical mark reader.
  • the translation apparatus 1 also includes an output 8 selected from one or more or any combination of the following: a video output device, a monitor, a display screen, an audio output device, a speaker, an earphone jack, an earphone, and a printing device.
  • the translation apparatus 1 includes a memory 9 for storing the database, the memory selected from one or more or any combination of the following: a semiconductor storage device, a ROM IC, flash memory, a hard disk, and magnetic tape.
  • the translation apparatus 1 is in the form of an integrated translation unit for use by a user.
  • the integrated translation unit includes the memory 9 as an onboard memory and is supplied with the database loaded on the onboard memory.
  • the translation apparatus 1 also includes a translation unit for use by a user, but the memory 9 on which the database 2 is stored is a remote memory.
  • the translation unit communicates with the remote memory 9 to access the database.
  • the translation unit is a portable translation device for personal use, and the database 2 is stored on the memory 9 of a remote computer server.
  • the translation device communicates with the remote computer server to access the database on the memory 9 of the remote server computer using any suitable communication means, such as LAN, WAN, Bluetooth, Wi-Fi, and 3G.
  • the translation unit has an onboard memory 9 and the translation unit downloads the database from the remote memory onto the onboard memory.
  • the translation unit is again a portable translation device for personal use, and the database 2 is again stored on the memory of a remote computer server.
  • the translation device communicates with the remote computer server to download the database from the memory of the remote server computer onto the onboard memory 9 of the translation device using any suitable communication means, such as LAN, WAN, Bluetooth, Wi-Fi, and 3G.
  • the translation unit performs the following steps: receive commands from a user; access the database 2 according to the commands to find a phrase 3 selected by the user for translation from a source language to at least one target language; and output the selected phrase 3 translated into the at least one target language.
  • the translation unit has an onboard memory 9 and the steps are in the form of an executable program stored on a remote memory, the translation unit downloading the executable program onto the onboard memory 9 to execute the program to perform the steps.
  • a smartphone can download the executable program described above as an application from an online application store. The smartphone can then execute the program to perform the steps above and take the role of the translation unit.
  • the same smartphone in the role of the translation unit can also download the database 2 from the same online application store or from some other remote computer server.
  • the executable program and the database 2 are combined to form part of one application which can be downloaded, thereby allowing the executable program and the database to be downloaded at the same time.
  • the database remains on the remote computer server.
  • the smartphone having downloaded the executable program, simply accesses the database 2 whilst it is still on the memory of the remote computer server. This allows the database 2 to be continually updated, and the smartphone, as the translation unit, will always access the latest version of the database.
  • a dedicated translation unit can be used instead of a smartphone.
  • the steps can be performed through the use of an executable program, and the executable program and/or the database can be updated by downloading new versions or a combined new version from a remote computer server. This can be done via any suitable communication means such as LAN, WAN, Bluetooth, Wi-Fi, and 3G.
  • another aspect of the present invention provides a non-transitory computer-readable storage medium with a database stored thereon, the database containing phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.
  • the database in a preferred embodiment of this aspect is the database 2 described above.
  • a preferred embodiment of a further aspect of the invention provides non-transitory computer-readable storage medium with an executable program stored thereon, wherein the program instructs a translation unit to perform the following steps: receive commands from a user; access the database 2 described above according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and output the selected phrase translated into the at least one target language.
  • the present invention allows users to find desired phrases for translation much faster and much more efficiently than prior translation devices, especially in the context of natural conversations.
  • the database of the present invention allows the linking of phrases to follow natural conversational flow, where the selection of a desired phrase by a user leads directly to further phrases that are likely to be needed by the user next.
  • the present invention presents phrases to users predictively.
  • the present invention avoids the need to return to the same starting point or to go back up to a top-level of the database after finding a desired phrase in order to find the next desired phrase.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • General Health & Medical Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Machine Translation (AREA)

Abstract

The present invention provides a translation apparatus including a database of phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.

Description

Translation Apparatus Technical Field
The present invention relates to translation apparatuses, and in particular, but not limited to, translation apparatuses for translating phrases used in everyday conversation. The invention will be described in relation to portable personal translation devices, but is not limited to this particular use.
Background Art
Existing translation devices include those that translate phrases used in everyday conversation from a source language to a target language. These types of devices are particularly suited for personal use in situations that require the real-time translation of phrases, such as when a user engages in conversation with a person who does not share a common language with the user, and during the ensuing conversation.
However, the number of phrases and types of subject matter covered by existing translation devices is limited. This is because the more phrases a translation device covers, the more time is required to find a phrase. This causes great difficulties when a translation device is relied upon to find a particular phrase required in the context of and during the flow of a natural everyday conversation. The resulting delays interrupt conversational flow, especially in time-critical situations or when the situation changes during the conversation, and affect the accuracy and efficiency of communication between the parties involved. This may have detrimental effects in situations such as emergencies where slow communication or misunderstandings can have dire consequences.
Existing translation devices typically include a database of phrases. In some existing translation devices, the databases are simply linear databases and users simply scroll through the phrases until the desired phrase is found. Once a desired phrase is found, the user must return to scrolling through the phrases to find the next desired phrase. This becomes very time-consuming and inefficient with a large number of phrases.
In other existing translation devices, phrases are grouped under category or topic headings in the database. Users select the applicable topic heading and then scroll through the phrases under the selected topic heading until the desired phrase is found. However, once a desired phrase is found, the user must start from the original starting point again to find the next desired phrase. In particular, if the next desired phrase is not in the current topic heading, the user must again select a topic heading and then scroll though the phrases under the topic heading to find the next desired phrase. If phrases are grouped under levels of category or topic headings, then a user must drill down the levels to the desired group of phrases to find the desired phrase. If the next phrase is not in the current group, then the user must start from the same starting point again, that is, the top level of topic headings. Similar to the linear database example above, this is still very time-consuming and inefficient with a large number of phrases.
It is an object of the present invention to overcome or ameliorate at least one of the disadvantages of the prior art, or to provide a useful alternative.
Technical Problem
The present invention, in a first aspect, provides a translation apparatus including a database of phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.
Technical Solution
Preferably, the database includes scenarios, each scenario linked to a respective batch such that selection of a scenario accesses a respective batch. In some embodiments, at least one scenario is a member of a batch.
Preferably, the database includes one or more series of batches linked successively together whereby a primary member of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and one or more of the batches in a series has a secondary member that is linked to another batch other than the next batch in the series.
Preferably, the database includes top-level scenarios, each top-level scenario linked to a respective series of batches such that selection of a top-level scenario accesses the first batch in a respective series of batches.
Preferably, the translation apparatus includes a user input selected from one or more or any combination of the following: a keyboard, a switch, a computer mouse, a touchpad, a touch screen, an audio input device, a microphone, a scanning device, a barcode reader, a card reader, an optical input device, and an optical mark reader. Also preferably, the translation apparatus includes an output selected from one or more or any combination of the following: a video output device, a monitor, a display screen, an audio output device, a speaker, an earphone jack, an earphone, and a printing device. Preferably, the translation apparatus includes a memory for storing the database, the memory selected from one or more or any combination of the following: a semiconductor storage device, a ROM IC, flash memory, a hard disk, and magnetic tape.
In some embodiments, the translation apparatus is in the form of an integrated translation unit for use by a user, the integrated translation unit having an onboard memory and supplied with the database loaded on the onboard memory.
In other embodiments, the translation apparatus includes a translation unit for use by a user and the database is stored on a remote memory. In one embodiment, the translation unit communicates with the remote memory to access the database. In another embodiment, the translation unit has an onboard memory and the translation unit downloads the database from the remote memory onto the onboard memory.
Preferably, the translation unit performs the following steps: receive commands from a user; access the database according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and output the selected phrase translated into the at least one target language.
In some embodiments, the translation unit has an onboard memory and the steps are in the form of an executable program stored on a remote memory, the translation unit downloading the executable program onto the onboard memory to execute the program to perform the steps.
In a second aspect, the present invention provides a non-transitory computer-readable storage medium with a database stored thereon, the database containing phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.
Preferably, the database includes scenarios, each scenario linked to a respective batch such that selection of a scenario accesses a respective batch. In some embodiments, at least one scenario is a member of a batch.
Preferably, the database includes one or more series of batches linked successively together whereby a primary member of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and one or more of the batches in a series has a secondary member that is linked to another batch other than the next batch in the series.
Preferably, the database includes top-level scenarios, each top-level scenario linked to a respective series of batches such that selection of a top-level scenario accesses the first batch in a respective series of batches.
In a third aspect, the present invention provides a non-transitory computer-readable storage medium with an executable program stored thereon, wherein the program instructs a translation unit to perform the following steps: receive commands from a user; access the database described above according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and output the selected phrase translated into the at least one target language.
Description of Drawings
Preferred embodiments in accordance with the best mode of the present invention will now be described, by way of example only, with reference to the accompanying figures, in which:
Fig. 1 is a block diagram of a translation apparatus in accordance with a preferred embodiment of the present invention;
Fig. 2 is a schematic diagram of the network structure of phrases in the database of a translation apparatus in accordance with another preferred embodiment of the present invention;
Fig. 3 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with another preferred embodiment of the present invention;
Fig. 4 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with yet another preferred embodiment of the present invention;
Fig. 5 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with yet another preferred embodiment of the present invention;
Fig. 6 is a schematic diagram of part of the network structure of phrases in the database of a translation apparatus in accordance with a further preferred embodiment of the present invention;
Fig. 7 is a flow diagram of the operation of a translation apparatus in accordance with a preferred embodiment of the invention;
Fig. 8 is a table showing part of a database of a translation apparatus in accordance with a preferred embodiment of the invention; and
Fig. 9 is a table showing another part of the database of Fig. 8.
Best Mode
Referring to the figures, a translation apparatus 1 in accordance with a preferred embodiment of the present invention is provided. The translation apparatus 1 includes a database 2 of phrases 3 for translation from a source language to at least one target language. The phrases in the database 2 are members 4 of batches 5. At least one member 4 in a parent batch is linked to a child batch such that the child batch is accessible by selecting the at least one member. The parent and child batches can be any two of the batches. The same batch 5 can be a parent batch and a child batch at the same time. The term “parent” when referring to a batch as a “parent batch” simply means that a member 4 of the batch is linked to another batch, which is referred to as a “child” batch. This “child batch” can also be referred to as a “parent batch” if it too has a member 4 that is linked to another batch.
The database also includes scenarios 6, with each scenario linked to a respective batch 5 such that selection of a scenario accesses a respective batch. At least one scenario 6 is a member 4 of a batch 5.
A phrase 3 is a natural language expression typically used in conversation, such as “Where is the nearest drug store?” or “Are there any duty-free concessions?”, as shown in Fig. 6. Phrases 3 are the members 4 that are primarily required to be translated by users. A scenario 6 is a higher-level or general description of members of a batch, that is, the phrases or sub-scenarios in the batch. A scenario is linked to the batch whose members are generally described by the scenario name so that selection of the scenario accesses the related batch. A phrase or a scenario can be selected by a user either for translation or for accessing the batch linked to the scenario or, if any, the batch linked to the phrase.
For example, referring to Fig. 4, phrases “How do I get there?” and “Could you take me there?”, which are members 4d of the same batch 5d, relate to and are therefore linked to scenario 6c “Drug Store Directions”. Scenario “Drug Store Directions” 6c is itself a member of a batch 5c that includes other scenarios 6c “Opening Hours” and “Payment” as members. This batch 5c is in turn linked to scenario 6b “Drug Store”, a member of a batch 5b which includes other scenarios “First Aid” and “Payment” as members. Finally, batch 5b is linked to scenario 6a “Medical & Emergencies”.
Scenarios 6c “Drug Store Directions”, “Opening Hours”, and “Payment” can be seen as sub-scenarios to scenario 6b “Drug Store”. Likewise, scenarios 6b “Drug Store”, “First Aid”, and “Payment” can be looked upon as sub-scenarios to scenario 6a “Medical & Emergencies”. Scenario 6a “Medical & Emergencies” can be referred to as a “top-level” scenario. Also, a batch 5 can have a mix of phrases 3 and scenarios 6 as its members 4. For example, batch 5d includes the phrases 4d “How do I get there” and “Could you take me there” with the scenario 6d “Back to Drug Store”.
The batches can be described as going from a higher level batch 5b to progressively lower level batches 5c and 5d. As can be seen, higher level batches generally cover broader subject matter than lower level batches which cover more and more specific subject matter. The links going from higher level batches to lower level batches can be referred to as forward links and the links going from lower level batches to higher level batches can be referred to as backward links.
Batches 5b and 5c are parent and child batches, respectively, since a member 4b of batch 5b, namely scenario 6b “Drug Store”, links to batch 5c. Batch 5c is also a parent batch in relation to batch 5d which is a child batch in this instance, due to the linking of member 4c in the form of scenario 6c “Drug Store Directions” to batch 5d. However, since a member 4d of batch 5d, namely scenario 6d “Back to Drug Store”, links back to batch 5c, batch 5d is also a parent batch and batch 5c is also a child batch.
Referring to the more expanded generalized database as shown in Fig. 2, the database 2 includes one or more series of batches 5x, 5y, 5z linked successively together whereby a primary member 4’ of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and wherein one or more of the batches in a series has a secondary member 4’’ that is linked to another batch other than the next batch in the series.
As well as members 4 linking to batches 5 in the same series, as described in the examples above in relation to Fig. 4, members can also link to batches 5 in a different series. Also, just as the links in the same series of batches, as per the examples of Fig. 4, can be forward links and backward links, members 4 can also link with batches 5 in a different series with forward links or backward links.
Referring to Fig. 5, series of batches 5g, 5h, and 5i and series of batches 5p, 5q, and 5r are shown. In each series, a member 4 of one batch links to the next batch. However, member 4r of batch 5r, in the form of phrase 3r “My baby is not very well”, links to batch 5i, which is in a different series of batches to batch 5r. In another embodiment, a further member 4r, such as phrase 3r “Can you help me with my baby”, links with another batch 5s in the same series. In this case, phrase 3r “Can you help me with my baby” would be a primary member 4’, and phrase 3r “My baby is not very well” would be a secondary member 4’’.
Referring to Fig. 6, series of batches 5j, 5k, and 5l and series of batches 5t, 5u, and 5v are shown. In each series, a member 4 of one batch links to the next batch. However, member 4v of batch 5v, in the form of phrase 3v “Is there a pharmacy”, links to batch 5k, which is in a different series of batches to batch 5v. This represents a backward link. In another embodiment, a further member 4v, such as phrase 3v “Are there any duty-free concessions”, links with another batch 5w in the same series. In this case, phrase 3v “Are there any duty-free concessions” would be a primary member 4’, and phrase 3v “Is there a pharmacy” would be a secondary member 4’’.
As shown in Fig. 2, there are also user-defined batches 5f. These batches contain phrases 3f or scenarios 6f from the database 2 marked as favourites by a user. These can be linked to user-defined scenarios 6e that have descriptions defined by the user.
The database further includes top-level scenarios 6TL, with each top-level scenario linked to a respective series of batches 5x, 5y, 5z such that selection of a top-level scenario accesses the first batch in a respective series of batches. For example, in Fig. 5, top-level scenarios 6TL “Medical & Emergencies” and “Transportation” link to first batches 5g and 5p, respectively.
The database structure described above is particularly useful for organizing phrases in logical flows according to natural everyday conversation. The variety of linkages between phrases allows phrases that would follow each other in logical conversational flows to appear to the user in accordance with those logical flows or closely matching those logical flows. The organization of phrases in batches and the linkages can be informed by human conversational habits. These can be undertaken by interviewing or surveying people in relevant situations, such as travelers and business people, and analyzing the results to determine typical conversational flows in a variety of situations.
For example, referring to Fig. 6, a user at an airport looking for a pharmacy starts by selecting top-level scenario 6TL “Transportation” to access batch 5t with scenario members 6t “Air Travel” and “Travelling by Land”. The user then selects “Air Travel” to access batch 5u with scenario members 6u “Airport Shops Directions” and “Shopping”. The user then selects “Airport Shops Directions” to access batch 5v with phrase members 3v “Are there any duty-free concessions?” and “Is there a pharmacy?”. The user then selects the desired phrase “Is there a pharmacy?” for translation from the user’s source language to the target language of an airport staff member or a passer-by.
All of the phrases 3 in the database 2 are in multiple languages. All of the versions of the phrases 3 in the different languages are linked together so that when a phrase 3 in a source language is selected for translation to a target language, the version of the phrase 3 in the target language is provided for output.
After translation, batch 5k with scenario members 6k “Drug Store Directions”, “Opening Hours”, and “Payment” is accessed, which are scenarios that link to likely phrases that would be required in the next stage of the conversation, such as phrase member 3l “Where is the nearest drug store?” in batch 5l, which can be accessed by selecting scenario member 6k “Drug Store Directions” in batch 5k.
The above is an example of the operation of translation apparatus 1 which is generalized in the flow diagram shown in Fig. 7.
An example of a specific implementation of the database 2 is shown in Fig. 8, which is in the form of a table depicting part of the database 2. The first column lists the location IDs of members 4, each of which is in a parent batch 5. The second column lists the location IDs of the members of the child batch 5 linked to a member 4 in a parent batch 5, with the member of the parent batch listed in the first column on the same row of the table. As can be seen, the member 4 with the location ID “1010010002” in the first column is a member of a parent batch 5, and is linked to a child batch with members having location IDs “1010020002”, “1010020356”, “1010020666”, and “1010010002”. Likewise, the member 4 with the location ID “1010020002” in the first column is a member of a parent batch 5, and is linked to a child batch with members having location IDs “1010030002”, “1010030050”, “1010030187”, “1010030202”, “1010030261”, “1010030276”, “1010030283”, “1010030293”, and “1010010002”.
The third column lists the ID of the phrase or scenario having the location listed in the first column on the same row of the table. The member 4 with the location ID “1010010002” in the first column has the phrase or scenario ID “0001”. The fourth column lists the ID of the phrase or scenario having the location listed in the second column on the same row of the table. The member 4 with the location ID “1010020666” in the second column has the phrase or scenario ID “9652”.
In this particular implementation, the location IDs are coded so that the first three digits of the location ID identify the top-level scenario 6TL, the second three digits identify a column in a part of the database, and the last four digits identify a row in a part of the database.
Using the phrase or scenario ID, the actual phrase or scenario wording in multiple languages can be found in another part of the database. For example, Fig. 9 shows the part of the database where the actual phrase or scenario wording in multiple languages can be found for the phrases or scenarios with IDs “0001” and “0002”.
The translation apparatus includes a user input 7 selected from one or more or any combination of the following: a keyboard, a switch, a computer mouse, a touchpad, a touch screen, an audio input device, a microphone, a scanning device, a barcode reader, a card reader, an optical input device, and an optical mark reader. The translation apparatus 1 also includes an output 8 selected from one or more or any combination of the following: a video output device, a monitor, a display screen, an audio output device, a speaker, an earphone jack, an earphone, and a printing device. Further, the translation apparatus 1 includes a memory 9 for storing the database, the memory selected from one or more or any combination of the following: a semiconductor storage device, a ROM IC, flash memory, a hard disk, and magnetic tape.
In one embodiment, the translation apparatus 1 is in the form of an integrated translation unit for use by a user. The integrated translation unit includes the memory 9 as an onboard memory and is supplied with the database loaded on the onboard memory.
In another embodiment, the translation apparatus 1 also includes a translation unit for use by a user, but the memory 9 on which the database 2 is stored is a remote memory.
In one embodiment, the translation unit communicates with the remote memory 9 to access the database. For example, the translation unit is a portable translation device for personal use, and the database 2 is stored on the memory 9 of a remote computer server. The translation device communicates with the remote computer server to access the database on the memory 9 of the remote server computer using any suitable communication means, such as LAN, WAN, Bluetooth, Wi-Fi, and 3G.
In another embodiment, the translation unit has an onboard memory 9 and the translation unit downloads the database from the remote memory onto the onboard memory. For example, the translation unit is again a portable translation device for personal use, and the database 2 is again stored on the memory of a remote computer server. The translation device communicates with the remote computer server to download the database from the memory of the remote server computer onto the onboard memory 9 of the translation device using any suitable communication means, such as LAN, WAN, Bluetooth, Wi-Fi, and 3G.
The translation unit performs the following steps: receive commands from a user; access the database 2 according to the commands to find a phrase 3 selected by the user for translation from a source language to at least one target language; and output the selected phrase 3 translated into the at least one target language.
In one embodiment, the translation unit has an onboard memory 9 and the steps are in the form of an executable program stored on a remote memory, the translation unit downloading the executable program onto the onboard memory 9 to execute the program to perform the steps.
These embodiments allow the widely used smartphones of today to take the form of the translation unit. For example, a smartphone can download the executable program described above as an application from an online application store. The smartphone can then execute the program to perform the steps above and take the role of the translation unit.
In another example, the same smartphone in the role of the translation unit can also download the database 2 from the same online application store or from some other remote computer server. In one version, the executable program and the database 2 are combined to form part of one application which can be downloaded, thereby allowing the executable program and the database to be downloaded at the same time.
This also allows the supplier to update the executable program and/or the database at any time, and to place the updated version of the executable program and/or database onto the online application store or remote computer server for download to update the version of the executable program and/or database on the translation unit.
In a further example, the database remains on the remote computer server. The smartphone, having downloaded the executable program, simply accesses the database 2 whilst it is still on the memory of the remote computer server. This allows the database 2 to be continually updated, and the smartphone, as the translation unit, will always access the latest version of the database.
Of course, in these embodiments, instead of a smartphone, a dedicated translation unit can be used. Also, even with the integrated translation unit, the steps can be performed through the use of an executable program, and the executable program and/or the database can be updated by downloading new versions or a combined new version from a remote computer server. This can be done via any suitable communication means such as LAN, WAN, Bluetooth, Wi-Fi, and 3G.
In view of the embodiments described above, another aspect of the present invention provides a non-transitory computer-readable storage medium with a database stored thereon, the database containing phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches. The database in a preferred embodiment of this aspect is the database 2 described above.
A preferred embodiment of a further aspect of the invention provides non-transitory computer-readable storage medium with an executable program stored thereon, wherein the program instructs a translation unit to perform the following steps: receive commands from a user; access the database 2 described above according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and output the selected phrase translated into the at least one target language.
The present invention allows users to find desired phrases for translation much faster and much more efficiently than prior translation devices, especially in the context of natural conversations. In particular, the database of the present invention allows the linking of phrases to follow natural conversational flow, where the selection of a desired phrase by a user leads directly to further phrases that are likely to be needed by the user next. Thus, the present invention presents phrases to users predictively. The present invention avoids the need to return to the same starting point or to go back up to a top-level of the database after finding a desired phrase in order to find the next desired phrase.
Although the invention has been described with reference to specific examples, it will be appreciated by those skilled in the art that the invention can be embodied in many other forms. It will also be appreciated by those skilled in the art that the features of the various examples described can be combined in other combinations.

Claims (20)

  1. A translation apparatus including a database of phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.
  2. A translation apparatus according to claim 1 wherein the database includes scenarios, each scenario linked to a respective batch such that selection of a scenario accesses a respective batch.
  3. A translation apparatus according to claim 2 wherein at least one scenario is a member of a batch.
  4. A translation apparatus according to any one of claims 1 to 3 wherein the database includes one or more series of batches linked successively together whereby a primary member of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and wherein one or more of the batches in a series has a secondary member that is linked to another batch other than the next batch in the series.
  5. A translation apparatus according to claim 4 wherein the database includes top-level scenarios, each top-level scenario linked to a respective series of batches such that selection of a top-level scenario accesses the first batch in a respective series of batches.
  6. A translation apparatus according to any one of claims 1 to 5 including a user input selected from one or more or any combination of the following: a keyboard, a switch, a computer mouse, a touchpad, a touch screen, an audio input device, a microphone, a scanning device, a barcode reader, a card reader, an optical input device, and an optical mark reader.
  7. A translation apparatus according to any one of claims 1 to 6 including an output selected from one or more or any combination of the following: a video output device, a monitor, a display screen, an audio output device, a speaker, an earphone jack, an earphone, and a printing device.
  8. A translation apparatus according to any one of claims 1 to 7 including a memory for storing the database, the memory selected from one or more or any combination of the following: a semiconductor storage device, a ROM IC, flash memory, a hard disk, and magnetic tape.
  9. A translation apparatus according to any one of claims 1 to 8 in the form of an integrated translation unit for use by a user, the integrated translation unit having an onboard memory and supplied with the database loaded on the onboard memory.
  10. A translation apparatus according to any one of claims 1 to 8 including a translation unit for use by a user and wherein the database is stored on a remote memory.
  11. A translation apparatus according to claim 10 wherein the translation unit communicates with the remote memory to access the database.
  12. A translation apparatus according to claim 10 wherein the translation unit has an onboard memory and the translation unit downloads the database from the remote memory onto the onboard memory.
  13. A translation apparatus according to any one of claims 10 to 12 wherein the translation unit performs the following steps:
    receive commands from a user;
    access the database according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and
    output the selected phrase translated into the at least one target language.
  14. A translation apparatus according to claim 13 wherein the translation unit has an onboard memory and the steps are in the form of an executable program stored on a remote memory, the translation unit downloading the executable program onto the onboard memory to execute the program to perform the steps.
  15. A non-transitory computer-readable storage medium with a database stored thereon, the database containing phrases for translation from a source language to at least one target language, the phrases in the database being members of batches, at least one member in a parent batch being linked to a child batch such that the child batch is accessible by selecting the at least one member, the parent and child batches being any two of the batches.
  16. A non-transitory computer-readable storage medium according to claim 15 wherein the database includes scenarios, each scenario linked to a respective batch such that selection of a scenario accesses a respective batch.
  17. A non-transitory computer-readable storage medium according to claim 16 wherein at least one scenario is a member of a batch.
  18. A non-transitory computer-readable storage medium according to any one of claims 15 to 17 wherein the database includes one or more series of batches linked successively together whereby a primary member of each batch in a series is linked to the next batch in the series such that the next batch is accessible by selecting the primary member, and wherein one or more of the batches in a series has a secondary member that is linked to another batch other than the next batch in the series.
  19. A non-transitory computer-readable storage medium according to claim 18 wherein the database includes top-level scenarios, each top-level scenario linked to a respective series of batches such that selection of a top-level scenario accesses the first batch in a respective series of batches.
  20. A non-transitory computer-readable storage medium with an executable program stored thereon, wherein the program instructs a translation unit to perform the following steps:
    receive commands from a user;
    access the database of any one of claims 15 to 19 according to the commands to find a phrase selected by the user for translation from a source language to at least one target language; and
    output the selected phrase translated into the at least one target language.
PCT/CN2011/080710 2011-10-12 2011-10-12 Translation Apparatus WO2013053114A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2011/080710 WO2013053114A1 (en) 2011-10-12 2011-10-12 Translation Apparatus

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2011/080710 WO2013053114A1 (en) 2011-10-12 2011-10-12 Translation Apparatus

Publications (1)

Publication Number Publication Date
WO2013053114A1 true WO2013053114A1 (en) 2013-04-18

Family

ID=48081358

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2011/080710 WO2013053114A1 (en) 2011-10-12 2011-10-12 Translation Apparatus

Country Status (1)

Country Link
WO (1) WO2013053114A1 (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008243076A (en) * 2007-03-28 2008-10-09 Toshiba Corp Interpretation device, method and program
CN101526937A (en) * 2008-03-07 2009-09-09 株式会社东芝 A machine translating apparatus and method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2008243076A (en) * 2007-03-28 2008-10-09 Toshiba Corp Interpretation device, method and program
CN101526937A (en) * 2008-03-07 2009-09-09 株式会社东芝 A machine translating apparatus and method

Similar Documents

Publication Publication Date Title
CN107209781A (en) Use the contextual search of natural language
US8560570B2 (en) Automated propagation of non-conflicting queries in distributed databases
US20190130262A1 (en) Machine learning model repository
US20130151555A1 (en) Content reproduction device, control method for content reproduction device, control program, and recording medium
US10565984B2 (en) System and method for maintaining speech recognition dynamic dictionary
JP2016071439A (en) Translation method and translation system
US9626658B2 (en) System and method for generating a task list
US11100164B2 (en) Displaying videos based upon selectable inputs associated with tags
EP3107012A1 (en) Modifying search results based on context characteristics
US20110231792A1 (en) Virtual cultural attache
WO2017095113A1 (en) Method for providing translation service , and electronic device therefor
EP4029012A1 (en) Techniques for dialog processing using contextual data
US20180075020A1 (en) Date and Time Processing
CN107066226A (en) The audio input of field entries
US9892199B2 (en) Specialized virtual personal assistant setup
US9002772B2 (en) Scalable rule-based processing system with trigger rules and rule evaluator
WO2013053114A1 (en) Translation Apparatus
JP2015036886A (en) Information processing device, storage medium, and method
US9569530B2 (en) Extracting and mining of quote data across multiple languages
Tingle Communication breakdown in healthcare settings
US10963466B2 (en) Contextual associations for entity queries
JP6246626B2 (en) Speech synthesis system, speech conversion support device, and speech conversion support method
US20160092824A1 (en) System and method for case management information entry
CN110297825B (en) Data processing method, device, computer equipment and storage medium
JP2017211687A (en) Related document extraction device

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 11874085

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 11874085

Country of ref document: EP

Kind code of ref document: A1