WO1997028634A1 - Database access - Google Patents
Database access Download PDFInfo
- Publication number
- WO1997028634A1 WO1997028634A1 PCT/GB1997/000233 GB9700233W WO9728634A1 WO 1997028634 A1 WO1997028634 A1 WO 1997028634A1 GB 9700233 W GB9700233 W GB 9700233W WO 9728634 A1 WO9728634 A1 WO 9728634A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- representations
- vocabulary
- representation
- combination
- distinguishable
- Prior art date
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
- H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M3/00—Automatic or semi-automatic exchanges
- H04M3/42—Systems providing special services or facilities to subscribers
- H04M3/487—Arrangements for providing information services, e.g. recorded voice services or time announcements
- H04M3/493—Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals
- H04M3/4931—Directory assistance systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
- G10L2015/226—Procedures used during a speech recognition process, e.g. man-machine dialogue using non-speech characteristics
- G10L2015/228—Procedures used during a speech recognition process, e.g. man-machine dialogue using non-speech characteristics of application context
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M2201/00—Electronic components, circuits, software, systems or apparatus used in telephone systems
- H04M2201/40—Electronic components, circuits, software, systems or apparatus used in telephone systems using speech recognition
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99931—Database or file accessing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99931—Database or file accessing
- Y10S707/99933—Query processing, i.e. searching
- Y10S707/99936—Pattern matching access
Definitions
- the present invention relates to a database access particularly, though not exclusively, employing speech recognition input and synthesised speech output.
- International patent publication number WO94/14270 describes a mechanised directory enquiry system in which a caller is first prompted to speak the name of the city required. The word spoken is then recognised and the word with the highest confidence level is selected as being the word spoken by a user. The caller is then prompted to speak the name of the sought party. When a satisfactory confidence level is obtained, a database is accessed and the number articulated to the caller.
- the caller is prompted to spell all or part of the location or the name If more than one match between the spoken input and the database is found, the user is asked to confirm each match one by one until a confirmed match is found. If no such match can be located, the automatic processing is terminated.
- European patent application publication no. 433964 relates to a system which uses a text input. First of all an input word representing the surname is matched with the entries. If a match is found that is "comparable" but not exactly the same as the input, the initial characters of the input and the database entry are compared. If these match, a record of the data entry is made. The system then compares the required titles and the personal names required. The most likely entry is provided to the user
- US patent no. 5204894 relates to a personal electronic directory in which the names of the entries are stored in the user's voice and associated numbers are input using either a multitone (DTMF) telephone keypad or a spoken input.
- DTMF multitone
- the directory system compares the first word of the input with the stored words and provides all possibilities in sequence to the user until the user confirms one.
- the systems provide to the user identified entries in a database in a sequential manner until a user confirms the data entry as being that required.
- a database access apparatus comprising: (a) a database containing entries each comprising a plurality of fields which contain machine representations of items of information pertaining to the entry, the said representations forming a first vocabulary;
- announcement means responsive to machine representations falling within a second vocabulary of such representations to generate audio signals representing spoken announcements
- translation means defining a relationship between the first vocabulary and the second vocabulary and between the first vocabulary and the third vocabulary
- control means operable
- the included word output by the announcement means may be in any suitable form e.g. the included word may represent a whole word, a spelt word or alphanume ⁇ cs.
- the invention provides a method of speech recognition comprising
- Figure 1 is an Entity Relationship Diagram showing an example of translations between phonetic, spoken and database representations
- FIG. 2 is a block diagram of apparatus according to the invention.
- Figure 3 is a flow chart illustrating the operation of the apparatus of Fig. 1 ;
- Figure 3a is a flow chart illustrating an alternative operation of the apparatus of Figure 1 ;
- Figure 4 is a flow chart illustrating the process of identifying distinguishable tuples;
- Figure 5 is an Entity Relationship Diagram showing an example of the translations between phonetic, spoken, spelt and database representations.
- a voice interactive apparatus will be described, which generates questions to a user and recognises the user's responses in order to access the contents of the database.
- a database of names, addresses and telephone numbers will be used as an example. Firstly, however, some basic concepts will be discussed which will be of value in understanding the operation of the apparatus.
- the database will be supposed to contain a number of entries, with each entry containing a number of fields each containing an item of information about the entry; for example the forename, surname, location and telephone number of the person to whom the entry refers.
- a complete entry is thus a tuple as also is a smaller set of fields extracted from one entry; thus a set of forename/surname pairs taken from the example database forms a set of extracted duples.
- the items of information stored in the database fields may be in any convenient representation; generally this description will assume the use of a text representation such as, for the surname Jonson, character codes corresponding to the letters of the name, but with a stylised representation for some fields; for example one might, for geographical locations, identify several distinct places having the same name with different representations - e.g. Southend l , Southend2 and Southend3 for three place in England called Southend.
- a text representation such as, for the surname Jonson, character codes corresponding to the letters of the name, but with a stylised representation for some fields; for example one might, for geographical locations, identify several distinct places having the same name with different representations - e.g. Southend l , Southend2 and Southend3 for three place in England called Southend.
- the words used in the dialogue between the apparatus and the user to represent field contents are conceptually distinct from the database representations and represent for each field a spoken vocabulary. If the database representations are text then there will be some overlap between them and the spoken vocabulary, but even then it may be desired to take account of the fact that the user might use, to describe an item of information, a different word from that actually contained in the database field; that is, some words may be regarded as synonyms.
- Figure 1 is an "Entity Relationship Diagram", where we see a need for translation between representations as one moves from left to right or right to left.
- Box A represents a set of database entries.
- Box B represents a set of unique surnames, which have a 1 :many relationship with the entries -i.e. one surname may appear in many entries but one entry will contain only one surname.
- Boxes C, D and E correspond to sets of representations of forenames, towns and telephone numbers, where similar comments apply
- Box F represents the spoken vocabulary corresponding to forenames i.e. the set of all words that are permitted by the apparatus to be used to describe this field.
- Kesgrave A person living in Kesgrave might have his address recorded in the database either as Ipswich or as Kesgrave. Similarly an enquirer seeking the telephone number of such a person might give either name as the location.
- Ipswich and Kesgrave may be regarded as synonymous for the purposes of database retrieval Note however that this geographical aliasing is complex: Ipswich may be regarded as synonymous with another local village such as Foxhall, but Kesgrave and Foxhall are not synonymous because they are different places.
- Box H represents, for completeness, a spoken vocabulary for surnames, though there is probably little scope for synonyms for this field.
- Box J represents a pronunciation vocabulary for surnames, to take account of homophones and homonyms.
- the surname Smith is generally pronounced with a short "i” as in the English word “pith”
- the name Smythe is pronounced with a long "i” as in “lithe” .
- Smyth may be pronounced either way.
- Other instances of varying pronunciation may arise, for example due to variations in regional accents.
- “primary” and “may be” links are shown, for reasons to be explained later.
- Boxes K and L represent pronunciation vocabularies for forenames and geographical names respectively.
- FIG. 2 is a block diagram of an apparatus for conducting a dialogue.
- An audio signal input 1 is connected to a speech recogniser 2, whilst an audio signal output 3 is connected to a speech synthesiser 4
- a control unit in the form of a stored-program controlled processor 5 controls the operation of the recogniser and synthesiser and also has access to a program memory 6, a working memory (RAM) 7, a database 8, a spoken vocabulary translation table 9 and a pronunciation table 1 0.
- the audio inputs and outputs are connected for two-way communication - perhaps via a telephone line - with a user.
- the database 8 is assumed to contain telephone directory entries, as discussed above, in text form.
- the spoken vocabulary translation table 9 is a store containing word pairs consisting of a directory representation and a spoken vocabulary representation, e.g., for the Ipswich example, Database Spoken representation representation
- the translation table 9 has a separate area for each type of field and may be accessed by the processor 5 to determine the database representat ⁇ on(s) corresponding to a given vocabulary word and vice versa. If desired (or if the database representations are not in text form) all items may be translated.
- the pronunciation table 10 is a store containing a look-up table (and, if desired, a set of rules to reduce the number of entries in the look-up table) so that the processor 5 may access it (for synthesis purposes or for identifying homophones) to obtain, for a given spoken vocabulary word, a phonetic representation of one or more ways of pronouncing it, and, conversely (for recognition purposes), to obtain, for a given phonetic representation, one or more spoken vocabulary words which correspond to that pronunciation A separate area for each type of field may be desirable.
- the operation of the apparatus is illustrated in the flow-chart of Figure 3 - which is implemented as a program stored in the memory 6.
- the first steps involve the generation, using the synthesiser, of questions to the user, and recognition of the user's responses.
- the processor 5 sends to the synthesiser 4 commands instructing it to play announcements requesting the user to speak, respectively the surname, forename and town of the person whose telephone number he seeks.
- the processor sends to the recogniser 2 commands instructing it to recognise the user's responses by reference to phonetic vocabularies corresponding to those fields.
- the recogniser may access the translation table 9, 10 to determine the vocabularies to be used for each recognition step, or may internally store or generate its own vocabularies; in the latter case the vocabularies used must correspond to those determined by the table 9, 10 (and, if appropriate, the database) so that it can output only words included in the phonetic vocabulary.
- the recogniser is arranged so that it will produce as output, for each recognition step, as many phonetic representations as meet a predetermined criterion of similarity to the word actually spoken by the user. (The recogniser could of course perform a translation to spoken vocabulary representations, and many recognisers are capable of doing so) It is possible that the recogniser may indicate that the word actually spoken by the user is too dissimilar to any of the phonetic representations in Table 10 and indicate this to the processor 5. Preferably the recogniser also produces a "score" or confidence measure for each representation indicating the relative probability or likelihood of correspondence to the word actually spoken.
- the preliminary steps 100 - 1 10 will not be discussed further as they are described elsewhere; for example reference may be made to our co- pending International patent application no. PCT/GB/02524.
- steps will be described which involve the matching of a number of scored tuples against database entries. From these matching entries a scored set of unique (or distinguishable) tuples are derived which correspond to a different set of (possibly overlapping) fields to the tuples used for the match.
- step 1 10 the processor 5 has available to it, for each of the three fields, one or more phonetic representations deemed to have been recognised. What is required now is a translation to spoken vocabulary representations - i.e. the translation illustrated to the left of Figure 1 .
- the processor accesses the table 10 to determine, for each word, one or more corresponding spoken vocabulary representations, so that it now has three sets of spoken vocabulary representations, one for each field.
- the score for each spoken vocabulary representation is the score for the phonetic representation from which it was translated If two phonetic representations translate to the same vocabulary representation, the more confident of the two scores may be taken. This is a specific example of the generalised matching process described above where the matching set of singles are pronunciations and the derived set of singles are spoken vocabulary items
- step 1 1 4 the processor 5 now performs a translation to database representations - i.e the translation illustrated in the centre of Figure 1 - using the table 9 to determine, for each spoken representation, one or more corresponding database representations, so that it now has three sets of database representations Scores may be propagated as for the earlier translation
- the database representations represent a number of triples (the actual number being the product of the number of representations in each of the three sets)
- the score for a triple is typically the product of the scores of the individual representations of which it is composed.
- step 1 1 6 the processor generates a list of these triples and passes it to the database which returns a count K of the number of database entries corresponding to these triples If (step 1 1 8) this number is zero, then the processor in step 1 20 sends a command to the synthesiser to play an announcement to the effect that no entry has been found, and terminates the program (step 1 22) Alternatively other action may be taken such as transferring the user to a manual operator.
- step 1 24 the full entry tuples which matched at step 1 1 6 are retrieved in turn from the database to determine whether there are three or fewer distinguishable entries
- the tuples are retrieved in order of likelihood, most likely first. It is possible that more than one tuple may share the same score.
- step 1 26 the processor retrieves these entries from the database and forwards them to the synthesiser 4 which reads them to the user in confidence order, highest first, using the tables 9, 1 0 for translation from database representation to primary phonetic representation.
- the process enters an iterative confirmation phase in which an attempt is made to identify lists of extracted tuples which contain three or fewer distinguishable tuples, and to offer the tuples in turn to the user for confirmation.
- the tuples are the duple corresponding to the name (i.e. forename + surname), and the single corresponding to the town. Note that, although the case in this example, it is not in principle necessary that the constituent words of these tuples correspond to fields for which the user has already been asked.
- step 1 30 a check is made as to whether the name duples have already been offered for confirmation, on the first pass the answer will always be "no" , and at step 1 32 a list of extracted name duples is prepared from the list of triples.
- the name duples from the list are examined in similar fashion to that of the triples in step 1 24 to determine whether there are three or fewer distinguishable duples. (If desired the number of non-identical database representation duples in the list may be counted, and if this exceeds a predetermined limit, e.g. 30 the detailed examination process may be skipped (to step 144)).
- each of the scored duples are translated into a single primary phonetic representation and fed to the synthesiser in confidence order in step 136 so that the synthesiser speaks the question (e.g.) "is the name John Smith 7 please answer yes or no” one at a time with the recogniser forwarding the reply to the processor ( 138) for testing for "yes” or “no” . If the user replies "yes”, then, in step 140: (a) the surname and forename fields are marked “confirmed” so that further offering of them for confirmation is bypassed by the test at step 130;
- the process may then recommence from step 1 24. If a user replies "no" then the corresponding members in the list of triples are deleted. Which ones are deleted depends upon the defined relationship between the phonetic representations and the database representations, as chosen by the system designer For instance, if the user is asked "Is the name John Smith 7 " and the user replies "no", all members of the list of triples including John and Smith/Smyth/Smythe may be deleted or only those members including John and Smith/Smyth may be deleted, it having been decided by the system designer that Smythe is always pronounced differently to Smith or Smyth.
- step 142 If (step 142) the user has answered no to all the offered tuples, this is considered a failure and the process it terminated via steps 1 20 and 1 22.
- step 1434 If in the test at step 1 34 the number of distinguishable name duples is too large for confirmation, or at step 1 30 on a second or subsequent pass the name confirmation has already occurred, and assuming (step 144) the town name has not yet been offered for confirmation, then a town name confirmation process is commenced, comprising steps 1 46 to 1 54 which are in all respects analogous to the steps 1 32 to 1 42 already described. If these processes fail to reduce the number of distinguishable entries at the test 1 26 then the process eventually terminates with an announcement 1 56 that too many entries have been found for a response to be given. Alternatively, a further procedure may follow in which one or more further questions are asked (as in step 1 00) to obtain information on further fields.
- Two tuples are considered indistinguishable if every field of one tuple is indistinguishable (as defined above) from the corresponding field of the other tuple.
- Equally two representations are considered distinguishable if: (a) they are not identical; and (b) they do not translate to identical spoken vocabulary words (e.g. they are not synonyms or geographically confused); and
- the list is ordered by score; i.e. the tuple having the highest confidence is D( 1 ), the next D(2) and so on.
- the process to be described is illustrated in the flowchart of Figure 4 and involves taking the first tuple from the list, and comparing it with the tuple below it in the list to ascertain whether the two are distinguishable. If they are not, the tuple occupying the lower position is deleted from the list. This is repeated until all tuples have been examined. The same steps are then performed for the tuple now occupying the second position in the list, and so on; eventually every tuple remaining in the list is distinguishable from every other. If desired, the process may be terminated as soon as it is certain that the number of distinguishable tuples exceed that which can be handled by subsequent steps (i.e., in this example, 3).
- i points to a tuple in the list and j points to a tuple lower down the list.
- I is the number of tuples in the list.
- i is initialised to 1
- I is set to N
- D(i) is read from the database.
- Step 204 sets j to point to the following tuple and in step 206 D(j) is read.
- a field pointer m is then initialised to 1 in step 208 and this is followed by a loop in which each field of the two tuples is taken in turn.
- B is the number of such representations, i.e. A multiplied by the number of homophones.
- each of the phonetic representations p 1 (b) is compared with each of the representations p2(d) (i.e. BD comparisons in total) . If equality is not found in any of these comparisons, then the two tuples are considered distinguishable. If (step 226) j has not reached the last tuple in the list, it is incremented (228) prior to reading a further tuple in a repeat of step 206; otherwise the tupie pointer i is tested at step 230 as to whether it has reached the penultimate member of the list and either (if it has not) is incremented (232) prior to a return to step 202, or (if it has) the process ends. At this point, the list now contains only mutually distinguishable tuples - I in number - and thus the result k is set to I in step 233 prior to exit from this part of the process at step 234.
- the comparison at 21 8 indicates identity between one of the phonetic representations generated for one field of one tuple and one of the phonetic representations generated for the same field of the other tuple then it is necessary to increment m (step 236) and repeat steps 21 0 to 21 8 for a further field. If all fields of the two tuples have been compared and all are indistinguishable then this is recognised at step 238 and the tuples are deemed to be indistinguishable.
- the lower tuple D(j) is removed from the list and I is decremented so that it continues to represent the number of tuples remaining in the list (steps 240, 242) j is then tested at step 244 to determine whether it points beyond the end of the (now shortened) list and if not a further tuple is examined, continuing from step 206. Otherwise the process proceeds to step 230, already described.
- step 232 increments i to point to a tupie, it is known that there are at least i tuples which will not be removed from the list by step 240. Thus at this point ⁇ can be tested (step 246) to see if it has reached 3, and if so the process may be interrupted, k set to 4, and thence to the exit 234.
- ⁇ can be tested (step 246) to see if it has reached 3, and if so the process may be interrupted, k set to 4, and thence to the exit 234.
- step 1 the algorithm represents the execution of step 1 24, with the list at the conclusion of Figure 4 being used to access (from the database) the entries to be offered in step 1 28;
- step 1 32 the algorithm represents the execution of step 1 32, with the list at the conclusion of Figure 4 representing the list of name duples (in database representation) to be offered to the user in step 136;
- step 146 the algorithm represents the execution of step 146, with list at the conclusion of Figure 4 representing the list of towns to be offered to the user in step 1 50;
- step 140 the principle followed is that: where the user has confirmed a tuple (in this case a duple) which is one of a pair (or group) of tuples deemed indistinguishable, then this is considered to constitute confirmation also of the other tuple(s) of the pair or group. For example, if the list of name duples contains:
- step 1 32 only one entry (for example the first "Dave Smith") is offered to the user for confirmation in step 1 36.
- Which tuple is presented is determined according to a choice made by the system designer. However, if the user says “yes”, then in step 140, all tuples containing "Dave Smith” and all tuples containing "David Smyth" are retained.
- each field p of the confirmed duple in phonetic representation i.e. the one generated in step 1 36
- mappings used to IsP ⁇ ma ⁇ lySpoken/ output 146 decide on IsPrima ⁇ lyPronounced
- the processor 5 For all entries found the processor 5 then examines each representation contained in the selected field to identify distinguishable ones of those combinations. The distinguishable entries may then be presented to the user for confirmation. To achieve this, the processor 5, with reference to the spoken vocabulary store 1 0 and according to the defined relationships, translates the identified database representations into spoken vocabulary via the "may be spoken" route. Thus all entries of Dave are translated to "Dave” and “David”, all entries for David are translated to “Dave” or “David”, all entries for Mave are translated to "Mave” or “Mavis" and all entries of Mavis are translated to "Mave” or “Mavis”. The processor 5 then translates the spoken vocabulary representations into phonetic representations ("may be pronounced") with reference to store 1 0.
- phonetic representations which represent how "Dave”, “David”, “Mave” and “Mavis” are pronounced are determined as D Al V, D Al V I D, D AA V I D, M Al V and M Al VI S. These phonetic representations are then examined to identify distinguishable ones. For example, Dave and David are indistinguishable because they share at least one common pronunciation. However, Mave and Dave are distinguishable because they do not share any common phonetic representation. If two database representations are found to be indistinguishable, one of the representations is maintained and the other is discarded e.g.
- Figure 1 shows a situation in which the vocabulary of the announcement means e.g. the speech synthesiser is the same as the vocabulary of the input means e.g. the speech recogniser.
- spellings may also be used as an alternative input and/or confirmation medium to spoken forms.
- the techniques required for spelling are directly analogous to spoken forms.
- Figure 5 corresponds to Figure 1 with the inclusion of spelling (illustrated by box M) for town names (spelling may also be provided for surnames and forenames although, for simplicity, these mappings have not been shown in Figure 5) . Translations of "may be spelt" and “is primarily spelt" must be provided in addition to the spoken recognition.
- spoken or spelt input or output is not essential - since the considerations concerning the offering and confirming still arise.
- keypad input could be used, which has ambiguity problems owing to the allocation of more than one letter to each button of a telephone keypad.
- a further vocabulary - of keypad input codes - is required, with "May be keyed ." translations analogous to the pronunciation spelling translations described above.
- the machine representations of the input vocabulary and the database may be generated according to the same technique, for instance the database entries may be stored in text form and the input also be in text form, with the machine representations of the database and the output being generated according to a different technique e.g. a spoken output.
- Confusion may arise if a user is presented with an announcement which includes a synonym of the actual word said by the user. For instance, say a user asks for "Dave Smith” and the system geneates an output as follows- “Did you say David Smith 7 " In order to avoid this confusion, a check may be carried out to ensure that the word corresponding to the identified distinguishable database entry corresponds also to the word recognised by the input means
Abstract
Description
Claims
Priority Applications (7)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
NZ326441A NZ326441A (en) | 1996-01-31 | 1997-01-27 | Database access using speech recognition |
JP9527399A JP2000504510A (en) | 1996-01-31 | 1997-01-27 | Database access |
EP97901199A EP0878085B1 (en) | 1996-01-31 | 1997-01-27 | Database access |
DE69729277T DE69729277T2 (en) | 1996-01-31 | 1997-01-27 | DATABASE ACCESS |
AU36068/97A AU707248C (en) | 1996-01-31 | 1997-01-27 | Database access |
CA002244116A CA2244116C (en) | 1996-01-31 | 1997-01-27 | Database access |
NO983501A NO983501L (en) | 1996-01-31 | 1998-07-30 | Database access |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
GB9601925.2 | 1996-01-31 | ||
GBGB9601925.2A GB9601925D0 (en) | 1996-01-31 | 1996-01-31 | Database access |
US08/659,526 US5778344A (en) | 1996-01-31 | 1996-06-05 | Database access using data field translations to find unique database entries |
Publications (1)
Publication Number | Publication Date |
---|---|
WO1997028634A1 true WO1997028634A1 (en) | 1997-08-07 |
Family
ID=10787862
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/GB1997/000233 WO1997028634A1 (en) | 1996-01-31 | 1997-01-27 | Database access |
Country Status (12)
Country | Link |
---|---|
US (1) | US5778344A (en) |
EP (1) | EP0878085B1 (en) |
JP (1) | JP2000504510A (en) |
KR (1) | KR19990082252A (en) |
CN (1) | CN1121777C (en) |
CA (1) | CA2244116C (en) |
DE (1) | DE69729277T2 (en) |
GB (1) | GB9601925D0 (en) |
MX (1) | MX9806168A (en) |
NO (1) | NO983501L (en) |
NZ (1) | NZ326441A (en) |
WO (1) | WO1997028634A1 (en) |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2362746A (en) * | 2000-05-23 | 2001-11-28 | Vocalis Ltd | Data recognition and retrieval |
WO2004036887A1 (en) | 2002-10-16 | 2004-04-29 | Koninklijke Philips Electronics N.V. | Directory assistant method and apparatus |
EP1481328A1 (en) * | 2002-02-07 | 2004-12-01 | SAP Aktiengesellschaft | User interface and dynamic grammar in a multi-modal synchronization architecture |
US7603291B2 (en) | 2003-03-14 | 2009-10-13 | Sap Aktiengesellschaft | Multi-modal sales applications |
US8372418B2 (en) | 2007-07-20 | 2013-02-12 | Bayer Innovation Gmbh | Polymer composite film with biocide functionality |
US8383549B2 (en) | 2007-07-20 | 2013-02-26 | Bayer Cropscience Lp | Methods of increasing crop yield and controlling the growth of weeds using a polymer composite film |
CN107967916A (en) * | 2016-10-20 | 2018-04-27 | 谷歌有限责任公司 | Determine voice relation |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6301560B1 (en) * | 1998-01-05 | 2001-10-09 | Microsoft Corporation | Discrete speech recognition system with ballooning active grammar |
US6629069B1 (en) | 1998-07-21 | 2003-09-30 | British Telecommunications A Public Limited Company | Speech recognizer using database linking |
US6185530B1 (en) | 1998-08-14 | 2001-02-06 | International Business Machines Corporation | Apparatus and methods for identifying potential acoustic confusibility among words in a speech recognition system |
US6269335B1 (en) * | 1998-08-14 | 2001-07-31 | International Business Machines Corporation | Apparatus and methods for identifying homophones among words in a speech recognition system |
US6192337B1 (en) | 1998-08-14 | 2001-02-20 | International Business Machines Corporation | Apparatus and methods for rejecting confusible words during training associated with a speech recognition system |
JP2002532763A (en) * | 1998-12-17 | 2002-10-02 | コーニンクレッカ フィリップス エレクトロニクス エヌ ヴィ | Automatic inquiry system operated by voice |
GB2347307B (en) * | 1999-02-26 | 2003-08-13 | Mitel Inc | Dial by name feature for messaging system |
US6587818B2 (en) * | 1999-10-28 | 2003-07-01 | International Business Machines Corporation | System and method for resolving decoding ambiguity via dialog |
US7013276B2 (en) * | 2001-10-05 | 2006-03-14 | Comverse, Inc. | Method of assessing degree of acoustic confusability, and system therefor |
DE602004011753T2 (en) * | 2003-03-01 | 2009-02-05 | Coifman, Robert E. | Method and device for improving transcription accuracy in speech recognition |
JP3890326B2 (en) * | 2003-11-07 | 2007-03-07 | キヤノン株式会社 | Information processing apparatus, information processing method, recording medium, and program |
US20060167920A1 (en) * | 2005-01-25 | 2006-07-27 | Listdex Corporation | System and Method for Managing Large-Scale Databases |
DE102014114845A1 (en) * | 2014-10-14 | 2016-04-14 | Deutsche Telekom Ag | Method for interpreting automatic speech recognition |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1994014270A1 (en) * | 1992-12-17 | 1994-06-23 | Bell Atlantic Network Services, Inc. | Mechanized directory assistance |
Family Cites Families (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4831654A (en) * | 1985-09-09 | 1989-05-16 | Wang Laboratories, Inc. | Apparatus for making and editing dictionary entries in a text to speech conversion system |
GB8610809D0 (en) * | 1986-05-02 | 1986-06-11 | Smiths Industries Plc | Speech recognition apparatus |
US5027406A (en) * | 1988-12-06 | 1991-06-25 | Dragon Systems, Inc. | Method for interactive speech recognition and training |
AU631276B2 (en) * | 1989-12-22 | 1992-11-19 | Bull Hn Information Systems Inc. | Name resolution in a directory database |
US5251129A (en) * | 1990-08-21 | 1993-10-05 | General Electric Company | Method for automated morphological analysis of word structure |
US5204894A (en) * | 1990-11-09 | 1993-04-20 | Bell Atlantic Network Services, Inc. | Personal electronic directory |
US5454062A (en) * | 1991-03-27 | 1995-09-26 | Audio Navigation Systems, Inc. | Method for recognizing spoken words |
CA2088080C (en) * | 1992-04-02 | 1997-10-07 | Enrico Luigi Bocchieri | Automatic speech recognizer |
US5623578A (en) * | 1993-10-28 | 1997-04-22 | Lucent Technologies Inc. | Speech recognition system allows new vocabulary words to be added without requiring spoken samples of the words |
-
1996
- 1996-01-31 GB GBGB9601925.2A patent/GB9601925D0/en active Pending
- 1996-06-05 US US08/659,526 patent/US5778344A/en not_active Expired - Lifetime
-
1997
- 1997-01-27 NZ NZ326441A patent/NZ326441A/en unknown
- 1997-01-27 KR KR1019980705981A patent/KR19990082252A/en not_active Application Discontinuation
- 1997-01-27 EP EP97901199A patent/EP0878085B1/en not_active Expired - Lifetime
- 1997-01-27 CA CA002244116A patent/CA2244116C/en not_active Expired - Fee Related
- 1997-01-27 DE DE69729277T patent/DE69729277T2/en not_active Expired - Lifetime
- 1997-01-27 WO PCT/GB1997/000233 patent/WO1997028634A1/en not_active Application Discontinuation
- 1997-01-27 JP JP9527399A patent/JP2000504510A/en active Pending
- 1997-01-27 CN CN97191984A patent/CN1121777C/en not_active Expired - Lifetime
-
1998
- 1998-07-30 NO NO983501A patent/NO983501L/en not_active Application Discontinuation
- 1998-07-31 MX MX9806168A patent/MX9806168A/en unknown
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO1994014270A1 (en) * | 1992-12-17 | 1994-06-23 | Bell Atlantic Network Services, Inc. | Mechanized directory assistance |
Non-Patent Citations (1)
Title |
---|
D.J.ATTWATER ET AL: "ISSUES IN LARGE-VOCABULARY INTERACTIVE SPEECH SYSTEMS", BT TECHNOLOGY JOURNAL, vol. 14, no. 1, January 1996 (1996-01-01), IPSWICH SUFFOLK (GB), pages 177 - 186, XP000579339 * |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2362746A (en) * | 2000-05-23 | 2001-11-28 | Vocalis Ltd | Data recognition and retrieval |
EP1481328A1 (en) * | 2002-02-07 | 2004-12-01 | SAP Aktiengesellschaft | User interface and dynamic grammar in a multi-modal synchronization architecture |
EP1481328A4 (en) * | 2002-02-07 | 2005-07-20 | Sap Ag | User interface and dynamic grammar in a multi-modal synchronization architecture |
US7177814B2 (en) | 2002-02-07 | 2007-02-13 | Sap Aktiengesellschaft | Dynamic grammar for voice-enabled applications |
WO2004036887A1 (en) | 2002-10-16 | 2004-04-29 | Koninklijke Philips Electronics N.V. | Directory assistant method and apparatus |
US7603291B2 (en) | 2003-03-14 | 2009-10-13 | Sap Aktiengesellschaft | Multi-modal sales applications |
US8372418B2 (en) | 2007-07-20 | 2013-02-12 | Bayer Innovation Gmbh | Polymer composite film with biocide functionality |
US8383549B2 (en) | 2007-07-20 | 2013-02-26 | Bayer Cropscience Lp | Methods of increasing crop yield and controlling the growth of weeds using a polymer composite film |
CN107967916A (en) * | 2016-10-20 | 2018-04-27 | 谷歌有限责任公司 | Determine voice relation |
CN107967916B (en) * | 2016-10-20 | 2022-03-11 | 谷歌有限责任公司 | Determining phonetic relationships |
US11450313B2 (en) | 2016-10-20 | 2022-09-20 | Google Llc | Determining phonetic relationships |
Also Published As
Publication number | Publication date |
---|---|
KR19990082252A (en) | 1999-11-25 |
NO983501L (en) | 1998-09-30 |
DE69729277T2 (en) | 2005-06-02 |
CA2244116C (en) | 2001-11-27 |
CN1121777C (en) | 2003-09-17 |
NZ326441A (en) | 1999-09-29 |
EP0878085A1 (en) | 1998-11-18 |
CA2244116A1 (en) | 1997-08-07 |
MX9806168A (en) | 1998-10-31 |
US5778344A (en) | 1998-07-07 |
DE69729277D1 (en) | 2004-07-01 |
JP2000504510A (en) | 2000-04-11 |
CN1210643A (en) | 1999-03-10 |
EP0878085B1 (en) | 2004-05-26 |
AU3606897A (en) | 1997-08-22 |
NO983501D0 (en) | 1998-07-30 |
AU707248B2 (en) | 1999-07-08 |
GB9601925D0 (en) | 1996-04-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP0878085B1 (en) | Database access | |
US6996531B2 (en) | Automated database assistance using a telephone for a speech based or text based multimedia communication mode | |
KR100383352B1 (en) | Voice-operated service | |
US5987414A (en) | Method and apparatus for selecting a vocabulary sub-set from a speech recognition dictionary for use in real time automated directory assistance | |
US6324513B1 (en) | Spoken dialog system capable of performing natural interactive access | |
US8185539B1 (en) | Web site or directory search using speech recognition of letters | |
JPH10229449A (en) | Method and device for automatically generating vocabulary recognized talk out of registered item of telephone directory, and computer readable recording medium recording program element ordering computer to generate vocabulary recognized talk used in talk recognition system | |
JP2000032140A (en) | Robot hotel employee using speech recognition | |
WO2009149340A1 (en) | A system and method utilizing voice search to locate a procuct in stores from a phone | |
US7020612B2 (en) | Facility retrieval apparatus and method | |
US6629069B1 (en) | Speech recognizer using database linking | |
AU707248C (en) | Database access | |
JP3316826B2 (en) | Information guidance method and device | |
GB2304957A (en) | Voice-dialog system for automated output of information | |
CA2440463C (en) | Speech recognition | |
Williams | Dialogue Management in a mixed-initiative, cooperative, spoken language system | |
JPH09114493A (en) | Interaction controller | |
JP3576511B2 (en) | Voice interaction device | |
JPH10105190A (en) | Method performing inquiry to data base | |
KR930000809B1 (en) | Language translation system | |
JP2003029784A (en) | Method for determining entry of database | |
JPH09288495A (en) | Button specification and voice recognition jointly using type input method and device | |
KR0173914B1 (en) | Name Search Method in Voice Dialing System | |
JP2001013987A (en) | Method and apparatus for speech controller having improved phrase memory, use, conversion, transfer and recognition |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
WWE | Wipo information: entry into national phase |
Ref document number: 97191984.4 Country of ref document: CN |
|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AU CA CN JP KR MX NO NZ SG US |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): AT BE CH DE DK ES FI FR GB GR IE IT LU MC NL PT SE |
|
DFPE | Request for preliminary examination filed prior to expiration of 19th month from priority date (pct application filed before 20040101) | ||
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 1997901199 Country of ref document: EP |
|
WWE | Wipo information: entry into national phase |
Ref document number: 326441 Country of ref document: NZ |
|
ENP | Entry into the national phase |
Ref document number: 2244116 Country of ref document: CA Ref document number: 2244116 Country of ref document: CA Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 1019980705981 Country of ref document: KR |
|
WWE | Wipo information: entry into national phase |
Ref document number: PA/a/1998/006168 Country of ref document: MX |
|
WWP | Wipo information: published in national office |
Ref document number: 1997901199 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 1019980705981 Country of ref document: KR |
|
WWG | Wipo information: grant in national office |
Ref document number: 1997901199 Country of ref document: EP |
|
WWR | Wipo information: refused in national office |
Ref document number: 1019980705981 Country of ref document: KR |