CN104692198A - Elevator voice calling landing register device - Google Patents
Elevator voice calling landing register device Download PDFInfo
- Publication number
- CN104692198A CN104692198A CN201410427952.1A CN201410427952A CN104692198A CN 104692198 A CN104692198 A CN 104692198A CN 201410427952 A CN201410427952 A CN 201410427952A CN 104692198 A CN104692198 A CN 104692198A
- Authority
- CN
- China
- Prior art keywords
- call registration
- section
- confirmation
- elevator
- voice
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B1/00—Control systems of elevators in general
- B66B1/34—Details, e.g. call counting devices, data transmission from car to control system, devices giving information to the control system
- B66B1/46—Adaptations of switches or switchgear
- B66B1/468—Call registering systems
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B66—HOISTING; LIFTING; HAULING
- B66B—ELEVATORS; ESCALATORS OR MOVING WALKWAYS
- B66B2201/00—Aspects of control systems of elevators
- B66B2201/40—Details of the change of control mode
- B66B2201/46—Switches or switchgear
- B66B2201/4607—Call registering systems
- B66B2201/4638—Wherein the call is registered without making physical contact with the elevator system
- B66B2201/4646—Wherein the call is registered without making physical contact with the elevator system using voice recognition
Abstract
The invention provides an elevator voice calling landing register device with high operationality; the elevator voice calling landing register device needs no multiple-voice recognizer, uses a simple structure to improve voice recognition reliability, and can limit user voice recognition result confirmation; the elevator voice calling landing register device comprises a determination information storage portion used for storing determination information, wherein the determination information is used for confirmation of a destination floor recognized by the voice recognition portion for the user; the elevator voice calling landing register device also comprises a determination portion used for determining according to the recognized destination floor recognized by the voice recognition portion and the determination information.
Description
Technical field
The present invention relates to and enable user carry out the elevator voice call registration device of call registration by voice.
Background technology
Utilizing push-botton operation to carry out in the elevator of call registration, user have with visual disabilities, user holds luggage or very crowded situation is inferior, be sometimes difficult to carry out based on push-button call registration.Therefore, in the past, be known to the push-botton operation that do not need to utilize hand to carry out and the elevator voice call registration device of call registration can be carried out by phonetic entry.
In this elevator voice call registration device, when have input the voice that user sends voice input sections such as microphones, the voice that speech recognition section identification inputs, carry out the call registration to the destination floor that user says.But, in such elevator voice call registration device, there is the situation that destination floor that user says is different from the destination floor that speech recognition section identifies.
This is proposed to following elevator voice call registration device: the recognition result notifying speech recognition section to user, make when recognition result is identification by mistake, user can carry out correcting (for example, referring to patent documentation 1).
In addition, also proposed following elevator voice call registration device: the recognition result notifying speech recognition section to user, when have input cancellation vocabulary from voice input section in the specified time after notifying, cancel the up-to-date call registration to destination floor (for example, referring to patent documentation 2) being undertaken by speech recognition section identification registering.
In addition, also proposed following speech input device: the recognition result of multiple voice means of identification is compared, computing is carried out to fiduciary level, according to the fiduciary level calculated, any one (for example, referring to the patent documentation 3) in " registration ", " allowing user confirm ", " non-registration " these three kinds is exported to control part and notification unit.It, by the fiduciary level using multiple voice means of identification to improve speech recognition, thus, prevents the mistake of call to register, and defines speech recognition section notifies situation from recognition result to user, thus improve comfort feature.
Patent documentation 1: Japanese Unexamined Patent Publication 3-031176 publication
Patent documentation 2: International Publication WO2012/081082 publication
Patent documentation 3: Japanese Patent No. 3082618 publication
In the prior art, when speech recognition section identifies destination floor, destination floor to be notified to user, therefore, there is the problem of needed wait time, operability decline.In addition, register to prevent the mistake of call and limit the situation of confirmation user being carried out to voice identification result, needing to arrange multiple voice means of identification, there is the problem that device structure complicates.
Summary of the invention
The present invention completes to solve the problem, its object is to the elevator voice call registration device providing a kind of operability higher, it can improve the fiduciary level of speech recognition with succinct structure, and can limit the situation of confirmation user being carried out to voice identification result.
Be provided with in elevator voice call registration device of the present invention: judgement information storage part, it stores judgement information, and this judgement information is for determining whether confirmation user being carried out to the destination floor that speech recognition section identifies; And detection unit, its destination floor identified according to speech recognition section and judgement information carry out described judgement.
In the present invention, be provided with: judgement information storage part, it stores judgement information, this judgement information is for determining whether confirmation user being carried out to the destination floor that speech recognition section identifies, and detection unit, its destination floor identified according to speech recognition section and judgement information carry out described judgement.Therefore, without the need to the fiduciary level using multiple voice means of identification can improve speech recognition, can the situation of user being carried out to destination floor confirmation be limited, thus can operability be improved.
Accompanying drawing explanation
Fig. 1 is the block diagram of the structure that embodiment 1 is shown.
Fig. 2 is the diagram of circuit of embodiment 1.
Fig. 3 is the diagram of circuit of embodiment 2.
Fig. 4 is the block diagram of the structure that embodiment 3 is shown.
Fig. 5 is the diagram of circuit of embodiment 3.
Fig. 6 is the block diagram of the structure that embodiment 4 is shown.
Label declaration
10 voice call registration device, 11 voice input sections, 12 speech recognition sections, 13 identification information storage parts, 14 call registration sections, 14a detection unit, 15 judgement information storage parts, 16 notification units, 19 elevator controlling portions, 31 descending stair test sections
Detailed description of the invention
Below, with reference to accompanying drawing, embodiments of the present invention are described.
Embodiment 1.
Fig. 1 is the block diagram of the elevator voice call registration device that embodiments of the present invention 1 are shown.Fig. 2 is the diagram of circuit of embodiment 1.
In FIG, voice input section 11 is by receiving the input of the voice that the user of elevator sends and the device (such as, the microphone etc.) sound wave of the voice based on input being converted to electric signal is formed.In the car that voice input section 11 is arranged on the stop (not shown) of elevator and/or elevator, the user of (not shown) easily inputs the position of voice.The voice being input to voice input section 11 are converted to voice signal (Wave datas of voice) by this voice input section 11 and output to speech recognition section 12.
Speech recognition section 12 compares check the voice signal received from voice input section 11 and the identification information (Wave datas of voice) be stored in advance in identification information storage part 13, carry out voice recognition processing thus, and voice identification result is outputted to call registration section 14.
First speech recognition section 12 carries out analog/digital conversion (A/D conversion) to the voice signal (simulation) from voice input section 11, is converted into digital signal.Next, intercept out from the voice signal (digital signal) after this conversion between the articulating region as the phonological component by user's sounding.And then, carry out acoustical analysis, in this acoustical analysis, the voice signal (Wave datas of voice) between the articulating region intercept out this and the identification information (Wave datas of voice) be stored in advance in identification information storage part 13 compare and check.As identification information, store destination registration vocabulary speech data and cancel registration vocabulary speech data.Thus, according to the voice being input to voice input section 11 whether with the speech data of destination registration vocabulary and to cancel in the speech data of registration vocabulary some consistent carries out speech recognition.
Destination registration is that the vocabulary of the destination floor representing call registration (such as, represents " " one " [pronunciation " ichi " of 1 layer with the speech data of vocabulary; One] ", " " イ ッ カ イ " [pronunciation " ikkai "; One deck] " and represent " " two " [pronunciation " ni " of 2 layers; Two] ", " pronunciation " ニ カ イ " [" nikai "; Two layers] " etc.In addition, may not be the number of plies, but such as " メ イ Application Off ロ ア [pinao nobile] " etc. as floor title or be positioned at the title etc. of representative facility of this floor) the speech waveform data of destination registration vocabulary.
The speech data of cancellation registration vocabulary is specific vocabulary (such as, " キ ャ Application セ Le [cancellation] " or " the ト リ ケ シ [pronunciation " torikeshi " for cancelling call registration; Cancel] " etc.) speech waveform data.This cancellation registration vocabulary is set to the vocabulary different from destination registration vocabulary.In addition, be set to not be included in the destination registration vocabulary representing certain certain floor in vocabulary.
Call registration section 14, when being identified destination registration vocabulary by speech recognition section 12, determines whether to carry out recognition result confirmation (with or without confirmation) to user by detection unit 14a.Now, represent that the floor information of the destination registration vocabulary that speech recognition section 12 identifies is stored in call registration section 14.
When having confirmation in the result of determination determined by detection unit 14a, call registration section 14 by the message notice of floor that comprises represented by destination registration vocabulary that speech recognition section 12 identifies to notification unit 16.
Call registration section 14 has timer, when notification unit 16 starts to notify that in the rear specified time risen, speech recognition section 12 identifies cancellation registration vocabulary, deletes the floor information of preserving.
When notification unit 16 starts to notify that in the rear specified time risen, speech recognition section 12 identifies destination registration vocabulary, call registration section 14 determines whether by detection unit 14a confirmation user being carried out to new voice identification result again.Now, the information of the floor represented by the destination registration vocabulary newly identified newly is saved in call registration section 14.
When notification unit 16 starts to notify that in the rear specified time risen, speech recognition section 12 fails identify destination registration vocabulary and cancel registration vocabulary, call registration section 14 exports the call registration request signal of the floor represented for preserved floor information to elevator controlling portion 19.Then, the floor information of preserving is deleted.
When the result of determination determined by detection unit 14a is without confirmation, call registration section 14 exports the call registration request signal for the floor represented by preserved floor information to elevator controlling portion 19, and deletes the floor information of preserving.
Detection unit 14a is arranged in call registration section 14, extracting the presence or absence confirmation be associated with the floor represented by the destination registration vocabulary as voice identification result in its judgement information stored from judgement information storage part 15, carrying out the judgement with or without confirming.
Judgement information storage part 15 is previously stored with the judgement information used in the judgement of detection unit 14a.As judgement information, such as low for the reliability of the speech recognition being envisioned as speech recognition section 12 and identify many floors by mistake, will represent that carrying out the such presence or absence confirmation of " having confirmation " of voice identification result confirmation to user to associate with floor and store.On the other hand, reliability for the speech recognition being envisioned as speech recognition section 12 is high and identify few floor by mistake, will represent that not carrying out the such presence or absence confirmation of " without confirm " of voice identification result confirmation to user to associate with floor and store.
The reliability of the speech recognition of speech recognition section 12 is low and identify that many floors are such as " イ ッ カ イ (1 layer) " (pronunciations: ikkai) and " ニ カ イ (2 layers) " (pronunciation: nikai) by mistake.Short sound due to them vowel that is different or syllable is i-a-i, and thus the possibility of speech recognition section 12 generation mistake identification is higher.Equally, " go カ イ (5 layers) " (pronunciation: gokai) is o-a-i with the vowel of the syllable of " ロ ッ カ イ (6 layers) " (pronunciation: rokkai), and thus the possibility of speech recognition section 12 generation mistake identification is also higher.
Notification unit 16 is for containing voice identification result, the result of call registration and the cancellation result of aftermentioned call registration in the interior various information relevant to elevator situation to user's notification package.Notification unit 16 is made up of read out instruments such as sound ring device and Liquid Crystal Display such as loud speakers, and in the stop being arranged on the elevator being provided with voice input section 11 and/or car.
Notification unit 16, when receiving the information as the floor represented by the destination registration vocabulary of the recognition result of speech recognition section 12 from call registration section 14, is notified to user by voice broadcast service and/or picture display.As the particular content notified to user, such as, the text importing (such as, 2 ~ 3 seconds) on " registration X layer " (X is the numeral representing destination floor) such voice broadcast service and/or picture can be considered.User, according to the content of announcement from this notification unit 16, can know the floor of oneself wishing and whether be identified as destination floor.
The entirety that the motion of elevator is responsible for by elevator control gear 19 controls.In the control object of this elevator control gear 19, certainly also comprise the call registration of elevator and the response etc. to call registration.
In addition, speech recognition section 12 carry out to the Wave data of the voice be input in voice input section 11 with comparing between the identification information stored in identification information storage part 13 checks, the threshold value being judged to be the likelihood that they are consistent also can be set.Can be in this case, when not being the likelihood of more than threshold value, speech recognition section 12 exports any one recognition result that is all inconsistent, that namely can not identify in the vocabulary stored in the voice and identification information storage part 13 representing and be input in voice input section 11 to call registration section 14.
When the recognition result of speech recognition section 12 be can not identify, certainly, call registration section 14 does not carry out call registration.In addition, also can represent that recognition result is such information that can not identify by exporting from call registration section 14 to notification unit 16, notifying from notification unit 16 to user to represent speech recognition failure (such as, " can not identify ").
In addition, also can be configured to: at the call registration request signal exported from call registration section 14 ask the floor of registering in not dockable situation, exporting from elevator controlling portion 19 to notification unit 16 can not berthing signal.In this case, by notification unit 16, notify to user to represent the information that can not stop.
In addition, can set according to floor utilization and/or mansion purposes the judgement information be stored in advance in judgement information storage part 15.Such as, the floor that the specific users such as office floor use can being set to " having confirmation ", the dining room layer etc. that an unspecified number of user uses being set to " without confirming ".In addition, also can being, for office building, full floor is set as " having confirmation ", for common facilities etc., full floor being set as " without confirming ".
In addition, notification unit 16 is made up of read out instruments such as voice ring device and/or Liquid Crystal Display such as loud speakers, but, as notification unit 16, the illuminating state of the call registration button be arranged in car also can be used.That is, when have input voice and identify destination floor by speech recognition section 12, such as, can glimmer by making the call registration button of this destination floor, identifying destination floor to user's notice.Continue the flicker of this call registration button, until through specified time.In addition, when determining call registration when have passed through specified time, this call registration button is lighted.
Next, action is described.The elevator voice call registration device of present embodiment carries out action according to the flow process shown in Fig. 2.
In step S101, confirm, in voice input section 11, whether to there is the input of the voice that user sends.When there is not phonetic entry in voice input section 11, flow process is made to proceed to step S117 and terminate.On the other hand, when there is phonetic entry in voice input section 11, step S102 is entered.
In step s 102, speech recognition section 12 is carried out comparing between the Wave data of the voice be imported in voice input section 11 and the Wave data of the voice as the identification information stored in identification information storage part 13 and is checked.Carry out speech recognition thus, then enter step S103.
In step s 103, confirm whether speech recognition section 12 identifies destination registration vocabulary in step s 102.That is, confirm whether speech recognition section 12 identifies the destination floor of user in step s 102.When speech recognition section 12 does not identify destination floor in step s 102, flow process is made to proceed to step S117 and terminate.On the other hand, when identifying destination floor in step s 102, step S104 is entered.
In step S104, detection unit 14a carries out the up-to-date destination floor that speech recognition section 12 in step S102 or step S113 identifies to be checked with comparing between the judgement information stored in judgement information storage part 15.Carry out thus, with or without the judgement confirmed, then entering step S105.
In step S105, when the result of determination of detection unit 14a is without when confirming, enter step S106, when the result of determination of detection unit 14a is, when having confirmation, enter step S107.
In step s 106, call registration section 14 exports the call registration request signal for the destination floor identified in step s 102 to elevator control gear 19, and elevator control gear 19 carries out the call registration for this destination floor.Then, flow process is made to proceed to step S117 and terminate.
In step s 107, notices such as " registration X layers " that from call registration section 14 to notification unit 16 output packet containing the information of destination floor, notification unit 16 is carried out.Then, step S108 is entered.
In step S108, call registration section 14 starts timer, and the measurement of time opening, then enters step S109.
In step S109, when have passed through specified time after time opening measurement in step S108, enter step S110.On the other hand, when in step S108 the time opening measurement after not yet through specified time, enter step S112.
In step s 110, call registration section 14 exports the call registration request signal of the up-to-date destination floor identified for speech recognition section 12 in step S102 or step S113 to elevator control gear 19.Elevator control gear 19 carries out call registration according to call registration request signal.Then, step S111 is entered.
In step S111, after call registration section 14 finishes time measurement, flow process is made to proceed to step S117 and terminate.
In step S112, confirm the input that whether there are the voice that user sends in voice input section 11.When there is not phonetic entry in voice input section 11, turn back to step S109.On the other hand, when there is phonetic entry in voice input section 11, step S113 is entered.
In step S113, speech recognition section 12 is checked by carrying out comparing between the Wave data of the voice be imported in voice input section 11 and the speech data of the identification information stored in identification information storage part 13, carry out speech recognition, then enter step S114.
In step S114, when the recognition result of the speech recognition section 12 in step S113 is, when cancelling registration vocabulary (cancelling call registration keyword), make flow process proceed to step S117 through step S111 and terminate.On the other hand, when the recognition result of the speech recognition section 12 in step S113 is not, when cancelling registration vocabulary, enter step S115.
In step sl 15, when the recognition result of the speech recognition section 12 in step S113 is destination registration vocabulary, step S116 is entered.On the other hand, when the recognition result of the speech recognition section 12 in step S113 is not destination registration vocabulary, the voice be input in voice input section 11 are considered as the noise that can not identify, phonetic entry are set to invalid and turn back to step S109.Now, the time measurement started in step S108 is not reset, and make it proceed.
In step S116, call registration section 14, after finishing time measurement, turns back to step S104.Then, in step S104, the new destination floor that detection unit 14a carries out identifying in step S113, with or without the judgement confirmed, then enters step S105.
In addition, when turning back to S109 from S115, also can notify to represent that speech recognition section 12 fails to identify the information of vocabulary and/or reminds the information again voice input section 11 being inputted to voice from notification unit 16.Such as, broadcast such as " pardon " can be carried out from notification unit 16.
In addition, in the present embodiment, when when turning back to step S109 from step S115, the time measurement started in step S108 is not reset at, but, also can reset the time measured, then turn back to step S109.
In addition, in the present embodiment, when not identifying cancellation registration vocabulary in step S113, step S115 being entered from step S114, but, also can be configured to turn back to step S109 from step S114.That is, also can be configured to only cancel.Thus, even if when other utilization person continues to speak at the appointed time, before the confirmation of the voice identification result before not terminating, do not carry out next and judge.
In the call registration device of such elevator, according to recognition result and the judgement judgement information stored in information storage part 15 of speech recognition section 12, define the situation allowing user confirm the recognition result of speech recognition section 12.Therefore, compared with carrying out the situation of destination floor confirmation all the time, operability can be improved.
In addition, with carry out the mode of call registration all the time when speech recognition section 12 identifies destination floor compared with, the situation that the call by mistake identified is registered can be reduced.In addition, speech recognition section 12, without the need to using multiple speech recognition equipment, can form by succinct equipment the fiduciary level improving speech recognition.
Embodiment 2.
Fig. 3 is the diagram of circuit of present embodiment.In figure 3, the part having marked label identical with Fig. 2 is the part same or equivalent with it.
In identification with in information storage part 13, except destination registration vocabulary and cancellation registration vocabulary, also store the speech waveform data confirming to specify vocabulary (such as, " registration ", " confirmation ").
Confirm to specify vocabulary even if then say after saying destination registration vocabulary, speech recognition section 12 also can identify destination registration vocabulary and confirm to specify these both sides of vocabulary.
When being identified confirmation by speech recognition section 12 and specifying vocabulary, detection unit 14a, according to by the presence or absence confirmation confirming to specify vocabulary to specify, determines whether the confirmation of recognition result user being carried out to speech recognition section 12.Other structure is identical with embodiment 1.
Next, the action different from embodiment 1 is described.The elevator voice call registration device of present embodiment carries out action according to the flow process shown in Fig. 3.
When being judged as in step s 103 identifying destination floor in step s 102, enter step S201.
When being judged in step s 201 not identify confirmation appointment vocabulary in step s 102, enter step S104.On the other hand, when identifying confirmation appointment vocabulary in step s 102, due to by confirming to specify vocabulary to specify with or without confirmation, thus skipping the presence or absence confirmation judgement of the detection unit 14 in step S104 and entering step S105.
In step S116, call registration section 14, after finishing time measurement, enters step S202.
When being judged to not identify confirmation appointment vocabulary in step S202 in step S113, turn back to step S104.On the other hand, when identifying confirmation appointment vocabulary in step S113, step S105 is turned back to.
In step S105, speech recognition section 12 is used in the up-to-date confirmation identified in step S102 or step S113 and specifies vocabulary.When this confirmation specifies vocabulary to be vocabulary (such as " registration ") representing " without confirming ", flow process is made to proceed to step S117 through step S106 and terminate.On the other hand, when the confirmation identified in step S102 or step S113 specifies vocabulary to be vocabulary (such as " confirmation ") representing " having confirmation ", step S107 is entered.
In the call registration device of such elevator, user can specify the confirmation of the recognition result whether user being carried out to speech recognition section 12.Therefore, even if even if when exist correctly identify also to carry out the floor confirmed when or identify that more situation of also not carrying out the floor confirmed is inferior when existing by mistake, user can specify with or without confirmation, thus improves comfort feature.
In addition, also can be configured to, specify the confirmation whether carrying out the keyword of the judgement of detection unit 14a execution to specify vocabulary as user, even if said before destination registration vocabulary, also can be identified by speech recognition section 12.
In addition, also can be configured to, can by obtaining confirmation set information from storing with portable terminal devices such as confirming the card of the confirmation set information of specifying vocabulary suitable and key, set the confirmation of the recognition result whether user being carried out to speech recognition section 12.
Embodiment 3.
Fig. 4 is the block diagram of the elevator voice call registration device that present embodiment is shown.In the diagram, the part having marked label identical with Fig. 1 is the part same or equivalent with it.
In the present embodiment, call registration section 14, according to the result of determination of detection unit 14a, is rewritten the judgement information of judgement information storage part 15, is upgraded.
Call registration section 14, according to the result of determination of detection unit 14a, calculates the score of floor.In addition, call registration section 14, according to the score calculated, rewrites the judgement information of judgement information storage part 15.Such as, the ratio, the i.e. discrimination that calculate vocabulary that user the says number of times consistent with the destination registration vocabulary that speech recognition section 12 identifies are used as score, and the presence or absence confirmation judging to comprise in information is rewritten as: for discrimination as more than 90% floor as " without confirm ", for discrimination lower than 90% floor be " having confirmation ".
When being judged to be " having confirmation " at detection unit 14a and starting in notification unit 16 to notify not identify cancellation registration vocabulary or new destination registration vocabulary in the rear specified time risen, the floor used in the judgement just carried out calculates the score as the floor correctly identified by call registration section 14.
On the other hand, when being judged to be " having confirmation " at detection unit 14a and starting in notification unit 16 to notify to identify cancellation registration vocabulary in the rear specified time risen, the floor used in the judgement just carried out is calculated the score as the floor identified by mistake.
In addition, be judged to be " having confirmation " at detection unit 14a and in the specified time risen after starting in notification unit 16 to notify when identifying the destination registration vocabulary made new advances, the floor used in the judgement just carried out and the floor of destination registration lexical representation that newly identifies are calculated the score as the floor identified by mistake.
Therefore, even exist with or without confirmation be the floor of " without confirm " but by situation about identifying by mistake, and calculating is as the discrimination of score.Other structure is identical with embodiment 2.
Next, the action different from embodiment 2 is described.The elevator voice call registration device of present embodiment carries out action according to the flow process shown in Fig. 5.
In step s 106, call registration section 14 exports the call registration request signal to the up-to-date destination floor that speech recognition section 12 identifies to elevator controlling portion 19, then enter step S301.
In step S111, after call registration section 14 finishes time measurement, enter step S301.
In step S301, call registration section 14, according to the presence or absence confirmation must assigning to rewrite judgement information storage part 15 calculated, makes flow process enter step S117 and terminate.Therefore, when starting downstream, carry out by detection unit 14 judgement employing new revised judgement information.
In the call registration device of such elevator, can according to based on reality utilization obstacle, utilize the score of environment, the situation of carrying out destination floor confirmation is limited.Therefore, compared with only using the situation of the presence or absence confirmation preset, the fiduciary level of speech recognition can be improved.
In addition, also can use and to input and the aviation value being identified the consistent degree between the speech data of the judgement information stored the Wave data of the voice of destination floor and judgement information storage part 15 by identification part 12 is used as score from voice input section 11.In this case, consistent degree is output to call registration section 14 from speech recognition section 12.
In addition, also can use and do not carry out cancelling or retelling etc. and be used as score by the number of times that phonetic entry is carried out registering continuously.Namely, can be configured to: for not carrying out cancelling or again judging, continuous stipulated number (such as, 10 times) register the floor of call, to be rewritten as with or without confirmation " without confirming ", for the floor cancelled or the floor again judged, confirmation information is rewritten as " having confirmation ".
In addition, about the calculating of score, the result of determination of the stipulated number (such as, 100 times) before also can only using, or give the weighting based on the elapsed time.Thus, according to the acoustic environment around the car of nearest elevator, rewrite with or without confirmation.
In addition, the threshold value of score is not limited to 90% of the threshold value as present embodiment.
Embodiment 4.
Fig. 6 is the block diagram of the elevator voice call registration device that present embodiment is shown.In figure 6, the part having marked label identical with Fig. 1, Fig. 4 is the part same or equivalent with it.
In the present embodiment, be provided with descending stair test section 31, it can detect the situation of user from lift car descending stair.Descending stair test section 31 detects the descending stair of user from car according to the exporting change of the optical sensor being arranged on car gangway, and exports descending stair to elevator controlling portion 19 and have no signal.
Elevator controlling portion 19 stores the voice register information representing and carry out the floor of registering according to the call registration request signal received from call registration section 14 as destination floor, and itself and descending stair is had together with no signal and output to call registration section 14.Call registration section 14 has no signal according to this voice register information and descending stair, rewrites the presence or absence confirmation be included in judgement information.Other structure is identical with embodiment 3.
Next, the action different from embodiment 3 is described.The elevator voice call registration device of present embodiment carries out action according to the flow process shown in Fig. 5.
In step S301, when descending stair test section 31 is when utilizing voice not detected user's descending stair by the floor registering call, be judged as that speech recognition section 12 has carried out identifying by mistake, the presence or absence confirmation be associated with this floor is rewritten as " having confirmation ".
On the other hand, in step S301, when descending stair test section 31 is when utilizing voice to be detected user's descending stair by the floor registering call, being judged as that speech recognition section 12 correctly identifies, the presence or absence confirmation be associated being rewritten as " without confirming " with this floor.
In the call registration device of such elevator, detecting the descending stair of user is set with without confirmation, therefore, rewrites compared with the situation with or without confirmation with only basis based on the information of phonetic entry, confirmation can be determined whether more accurately, can prevent the mistake of call from registering.Therefore, it is possible to raising operability.
In addition, descending stair test section 31 also can be configured to detect the descending stair of user according to the change of weight in car.
In addition, after flow process proceeds to step S301, can car arrive at the destination floor and upgrade with or without confirmation before, start new technological process, carry out step S101.
In addition, be set to descending stair test section 31 to elevator controlling portion 19 output signal, but also can be configured to call registration section 14 output signal.In this case, make call registration section 14 memory call registration request signal and register the floor of call, and having no signal according to the descending stair received from descending stair test section 31, rewriting the presence or absence confirmation be associated with this floor.
Claims (5)
1. an elevator voice call registration device, is characterized in that, described elevator voice call registration device has:
Speech recognition section, it identifies according to being input to the instruction of the voice in voice input section to user, and exports the instruction identified;
Storage part, it stores judgement information, and this judgement information is for determining whether the confirmation described user being carried out to the instruction that described speech recognition section identifies;
Detection unit, its instruction identified according to described speech recognition section and described judgement information, determine whether to carry out described confirmation;
Notification unit, it is when described detection unit is judged to carry out described confirmation, is contained in described judgement the described instruction that uses in interior information to described user's notification package; And
Call registration section, it exports the call registration request signal of the instruction used in the judgement just carried out based on described detection unit to elevator controlling portion.
2. elevator voice call registration device according to claim 1, is characterized in that,
When described detection unit is judged to not carry out described confirmation or when described notification unit carry out notify after rise specified time in described speech recognition section newly do not identify instruction, described call registration section exports described call registration request signal to described elevator controlling portion.
3. elevator voice call registration device according to claim 1, is characterized in that,
Described elevator voice call registration device also has can specify the designating unit of whether carrying out described confirmation for described user.
4. elevator voice call registration device according to claim 1 and 2, is characterized in that,
Described judgement information changes according to the result of determination of described detection unit.
5. elevator voice call registration device according to claim 1 and 2, is characterized in that,
Described judgement information changes according to the output from descending stair test section, and wherein, described descending stair test section carries out detection and output signal to described user from the situation of lift car descending stair.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2013251576A JP2015107867A (en) | 2013-12-05 | 2013-12-05 | Elevator voice call registration device |
| JP2013-251576 | 2013-12-05 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104692198A true CN104692198A (en) | 2015-06-10 |
| CN104692198B CN104692198B (en) | 2017-04-12 |
Family
ID=53339828
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410427952.1A Expired - Fee Related CN104692198B (en) | 2013-12-05 | 2014-08-27 | Elevator voice calling landing register device |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP2015107867A (en) |
| CN (1) | CN104692198B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018072319A1 (en) * | 2016-10-20 | 2018-04-26 | 深圳市元征科技股份有限公司 | Method for controlling elevator with voice and elevator control device |
| CN108584573A (en) * | 2018-05-07 | 2018-09-28 | 郑州科技学院 | A kind of Cargo Lift control system of artificial intelligence |
| CN108792856A (en) * | 2018-08-01 | 2018-11-13 | 迅达（中国）电梯有限公司 | elevator call system |
| CN109353902A (en) * | 2018-12-05 | 2019-02-19 | 广东卓梅尼技术股份有限公司 | Voice call method and system |
| CN109650199A (en) * | 2018-12-27 | 2019-04-19 | 日立电梯（中国）有限公司 | A kind of elevator voice operating method |
| CN110950201A (en) * | 2020-02-26 | 2020-04-03 | 菱王电梯股份有限公司 | Call registration method based on voice comparison and call registration device thereof |
| CN112093602A (en) * | 2020-11-13 | 2020-12-18 | 北京声智科技有限公司 | Elevator voice control method and device |
| CN112820271A (en) * | 2021-01-09 | 2021-05-18 | 江苏威尔曼科技有限公司 | Voice calling landing system with intelligent recognition function |
| CN112850391A (en) * | 2021-01-09 | 2021-05-28 | 江苏威尔曼科技有限公司 | Elevator voice calling landing registration device |
| CN113979249A (en) * | 2021-10-28 | 2022-01-28 | 上海三菱电梯有限公司 | Elevator voice control system and method |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112850390A (en) * | 2021-01-09 | 2021-05-28 | 江苏威尔曼科技有限公司 | Call registration method based on voice comparison |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0331176A (en) * | 1989-06-29 | 1991-02-08 | Toshiba Corp | Information input device for elevator |
| JPH08259124A (en) * | 1995-03-28 | 1996-10-08 | Fujitec Co Ltd | Voice recognizing device for elevator |
| JPH08259125A (en) * | 1995-03-28 | 1996-10-08 | Fujitec Co Ltd | Voice inputting device for elevator |
| CN103221325A (en) * | 2010-12-14 | 2013-07-24 | 三菱电机株式会社 | Call registration device of elevator |
-
2013
- 2013-12-05 JP JP2013251576A patent/JP2015107867A/en active Pending
-
2014
- 2014-08-27 CN CN201410427952.1A patent/CN104692198B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0331176A (en) * | 1989-06-29 | 1991-02-08 | Toshiba Corp | Information input device for elevator |
| JPH08259124A (en) * | 1995-03-28 | 1996-10-08 | Fujitec Co Ltd | Voice recognizing device for elevator |
| JPH08259125A (en) * | 1995-03-28 | 1996-10-08 | Fujitec Co Ltd | Voice inputting device for elevator |
| CN103221325A (en) * | 2010-12-14 | 2013-07-24 | 三菱电机株式会社 | Call registration device of elevator |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2018072319A1 (en) * | 2016-10-20 | 2018-04-26 | 深圳市元征科技股份有限公司 | Method for controlling elevator with voice and elevator control device |
| CN108584573A (en) * | 2018-05-07 | 2018-09-28 | 郑州科技学院 | A kind of Cargo Lift control system of artificial intelligence |
| CN108792856A (en) * | 2018-08-01 | 2018-11-13 | 迅达（中国）电梯有限公司 | elevator call system |
| CN109353902A (en) * | 2018-12-05 | 2019-02-19 | 广东卓梅尼技术股份有限公司 | Voice call method and system |
| CN109650199A (en) * | 2018-12-27 | 2019-04-19 | 日立电梯（中国）有限公司 | A kind of elevator voice operating method |
| CN110950201A (en) * | 2020-02-26 | 2020-04-03 | 菱王电梯股份有限公司 | Call registration method based on voice comparison and call registration device thereof |
| CN112093602A (en) * | 2020-11-13 | 2020-12-18 | 北京声智科技有限公司 | Elevator voice control method and device |
| CN112093602B (en) * | 2020-11-13 | 2021-11-02 | 北京声智科技有限公司 | Elevator voice control method and device |
| CN112820271A (en) * | 2021-01-09 | 2021-05-18 | 江苏威尔曼科技有限公司 | Voice calling landing system with intelligent recognition function |
| CN112850391A (en) * | 2021-01-09 | 2021-05-28 | 江苏威尔曼科技有限公司 | Elevator voice calling landing registration device |
| CN113979249A (en) * | 2021-10-28 | 2022-01-28 | 上海三菱电梯有限公司 | Elevator voice control system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2015107867A (en) | 2015-06-11 |
| CN104692198B (en) | 2017-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104692198A (en) | Elevator voice calling landing register device | |
| JP5601419B2 (en) | Elevator call registration device | |
| JP5626362B2 (en) | Elevator call registration device | |
| JP6502249B2 (en) | Speech recognition method and speech recognition apparatus | |
| JP5274569B2 (en) | Elevator destination floor registration device | |
| KR20190001434A (en) | System and device for selecting a speech recognition model | |
| US20220139377A1 (en) | Method for user voice input processing and electronic device supporting same | |
| KR20190090424A (en) | Method for responding user speech and electronic device supporting the same | |
| JPH0373775A (en) | Information input device for elevator | |
| JP2002128404A (en) | Device for registering call for elevator | |
| JP5545362B2 (en) | Elevator call registration device | |
| JP2011002534A (en) | Speech recognition device | |
| CN102892697B (en) | Elevator call registration device | |
| US20180350360A1 (en) | Provide non-obtrusive output | |
| JP2010054897A (en) | Voice recognition device, voice recognition program, reception device and reception program | |
| JP2807241B2 (en) | Voice recognition device | |
| JP3764302B2 (en) | Voice recognition device | |
| JP6335361B1 (en) | Elevator destination floor registration method | |
| JPH08259125A (en) | Voice inputting device for elevator | |
| KR200271438Y1 (en) | An apparatus for controling service of elevator by recognize voice | |
| JPH03293274A (en) | Voice register device for elevator | |
| JP2005084589A (en) | Voice recognition device | |
| JPH02147577A (en) | Elevator calling registration device | |
| KR20230091479A (en) | Elevator voice recognition operation control device for the visually impaired | |
| JP2009109535A (en) | Voice recognizer |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20170412 Termination date: 20200827 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |