KR940008345B1 - Voice controlled pager - Google Patents

Abstract

No content.

Description

Voice controlled pager, and its programming method

Subscribers to paging communication services continue to demand high performance and sophisticated features. On the other hand, these demands conflict with important requirements such as miniaturized size and long battery life. Modern phasers are designed to minimize their physical size. Thus, since these features typically require control or operation by the subscriber, it is difficult to integrate the improved features with conventional pagers, and the size of the pagers makes the addition of control switches or caps impractical.

Speech recognition technology is known to impart superior features in many products. Such speech recognition technology can be used to perform full or partial control as a voice command in a situation where hand-on control is impossible. The speaker dependent speech recognition device is designed to respond to a specific individual. Such devices store, for example, speaker specific parameters such as vocal tract response high periods, short term speech power, and amplitude changes in the short term speech spectrum.

Speech recognition causes some problems when applied to paging communication systems. One of them is the memory problem of speech recognition information (commonly referred to as "code book") in the pager. Currently, pagers are typically provided with subscriber specific information such as, for example, confirmation codes. On the other hand, this type of information can be easily generated and stored in the pager. Thus, if the pager is defective, the service company can easily provide a replacement pager to the subscriber. On the other hand, the generation and movement of the voice recognition codebook may not be the basic matter such as the confirmation code movement.

In some situations, it may be desirable to have a single pager answer more than one subscriber to a voice command. A typical situation may exist in a company setting where a pager is temporarily provided to an individual who has to respond to some emergency or critical situation. In other situations, a single pager may be shared, for example, during work shifts (ie, 8 hours and 3 shifts), eg between maintenance personnel operations. Thus, in order to provide voice control, a codebook for each individual must be stored in the pager. On the other hand, the large-capacity codebook storage device can greatly increase the price of the pager. In addition, the cost of maintaining the codebook is also significantly increased because the pager must be updated with each agent's change. Moreover, the addition of several memory devices or memory printed circuit boards can increase the size of the pager, which is inconvenient for the consuming public. This is of particular interest to paging subscribers. It is also possible to use speaker independent technology as an alternative, but speaker independent devices are typically more complex, less precise and consume more energy.

One solution to this problem is to store only a small number of codebooks, thus reducing the amount of memory required. This approach, on the other hand, is not appropriate for changing situations, such as fluctuations in personnel work time or a defect in the pager itself. Another approach could be to have each pager just embed enough memory in a single codebook (the codebook must be retrained for each subscriber). The speaker dependent sonic perception device must be "trained" to respond to each individual. The training is accomplished by causing the individual to continue word control several times until the speech recognition device is "trained" to recognize words spoken by an individual. On the other hand, this training takes time, and the addition of circuits that provide training capabilities for each pager may negatively impact pager price and size. Thus, there is a technical need to allow a pager to be programmed to respond to a subscriber's voice command in an effective and organized manner.

Summary of the Invention

Briefly, according to the first embodiment of the present invention, an individual who wants to operate a pager via a voice command sends a message requesting access to a specific codebook to a storage device of the voice recognition codebook. The storage device responds by sending the personal codebook to at least one pager, and the pager stores the codebook. The pager then responds to the voice command of the individual. In this way, the pager can be quickly reprogrammed.

According to a second embodiment of the invention, each person operating a pager in a paging communication system is provided with a codebook module (or card) which comprises at least suitable memory means in which the subscriber's codebook is stored. By inserting a codebook module (or card) into the pager, the pager can be at least partially controlled by the voice command of the subscriber. The codebook module may be easily removed so that the communication device can be quickly reprogrammed by inserting the codebook of another subscriber. The codebook module may also operate as a "key", which "locks" the pager upon removal of the codebook module, thereby inhibiting improper operation.

Detailed description of the preferred embodiment

1 shows a paging communication system 10 of the present invention. As shown, the paging communication system 10 includes at least one central store 12, which may communicate with multiple pagers (subscriber units) 14. As used herein, a pager or subscriber unit typically includes a personal, portable device designed to be carried by a person, and the term subscriber refers to any individual who uses or operates the pager or subscriber unit.

Typically, the reservoir 12 includes a transmitter (or a transceiver for an acknowledgment-back pager) 16, which is connected to the paging controller 18. Paging controller 18 includes appropriate address encoders and decoders, memory, processing facilities, real time clocks, and other related functions well known in paging technology. In addition, the paging controller 18 can access the database 20 of speech recognition information (codebook) and can route to the transmitter 16 for transmission to at least one phaser 14. The reservoir 12 also includes an interface 22 where communication with the outside world is received from the paging terminal and the telephone via a common public switched telephone network (PSTN) 24. As is known in the art, the paging terminal includes an input device through which messages and pager addresses are sent to the paging controller 18. Preferably, the paging controller includes a video terminal with an alphanumeric keypad (or equivalent) that allows for message entry. Optionally, one or more paging terminals 26 may be directly connected to the controller 18.

The operation of the paging communication system 10 shown in FIG. 1 shows that the pager 14 can receive messages in any of several message formats or signaling formats. Such message formats include, but are not limited to, the Golay Consecutive Code (GSC) and the British Post Apis Paging System Format (POCSAG). While only two message formats are mentioned, the present invention should consider the use of any suitable message format or signaling.

According to the first embodiment of the invention, all codebooks required for all personnel using the paging system 10 are contained in the center of the repository 12. The central repository is preferably associated with a facility that can be used to provide "education" of the subscriber unit. In this way, a new paging subscriber may create an agent codebook by training a pager (or simulated equivalent, such as a pager) at a central facility. The centralized storage of codebooks facilitates the maintenance and maintenance of codebook libraries. Instead, the storage device can be distributed among several locations within a given terrain area. On the other hand, this approach may require more stringent controls and adjustments to properly maintain the codebook library.

Each pager 14 contains enough codebook memory to store codebooks for at least one individual. Thus, before an individual uses the pager, his (or her) codebook must be moved (down-loaded) from the central repository 12 to the pager 14. Preferably, an individual who wishes to operate pager 14 may forward a message from the codebook library to the central repository 12 requesting confirmation and request of a particular codebook. The present invention contemplates some alternatives for generating a codebook request message. As a first method of generating a codebook request, a confirmation code may be entered on the keypad (or equivalent) of the paging terminal 26 and sent to the central repository 12. Instead, programming the pager 14 with a codebook can be accomplished by contacting the paging controller 18 via the PSTN 24 using any conventional telephone. In this way, codebook confirmation codes may be sent, for example, using DTMF phone tones, which may respond by downloading a particular codebook to a particular pager 14. In another embodiment, the paging controller 18 may use a limited variant of speaker independent speech recognition technology, which may be used to generate codebook download requests.

Finally, if the pager 14 includes a transmitter that enables recognition of the message, the codebook download request may be generated by sending an acknowledgment code from the pager to the paging transceiver. Upon receiving an unwanted acknowledgment code (ie, an acknowledgment code for which no message was previously sent), the paging transceiver may send the codebook to the request pager.

Once the codebook is stored in the pager 14, the subscriber can control the pager's operation in part or in full by voice commands. Instead, voice commands can be used in conjunction with manually entered commands (or vice versa) to perform the desired function. Whenever another person wants to use the pager (e.g., in case of job movement changes), the codebook for the new subscriber can be downloaded quickly, as mentioned above. Optionally, the subscriber's memory may be extended (within size and price constraints) to maintain a last limited number of end use codebooks in a first-in last out manner. For example, the codebook of the current user and the two previous subscribers can be maintained in the subscriber unit. In this way, no reprogramming of the subscriber unit will be required if the same three people operating the subscriber during the three-motion move are repeated every day.

In another aspect of the first embodiment, the present invention contemplates that the central repository 12 may include a database of subscribers and associated pages as they normally operate. At each movement change, the central repository 12 may automatically download the codebook for the next subscriber based on the time of day (or other appropriate parameter). In case of personnel changes due to illness or inefficiency, the supervisor can enter a new name into the roster (database) and the individual's codebook can be automatically downloaded to the subscriber unit. This approach has the advantage of minimizing over-the-air requests for new codebooks in situations where subscriber changes occur unexpectedly.

Referring to FIG. 2, a pager 14 'is shown to which the codebook module 30 is movably connected. The phaser 14 'is a shirt-pocket type with a display area 46 integrated with a holding or holding clip, and a microphone 48 for receiving voice commands. As shown, the phaser 14 'does not have any control (button or switch) commonly used with modern phasers. Instead, some buttons may be included if only partial voice control is implemented.

According to a second embodiment of the present invention, a codebook module (or a card depending on the size of the pager) is provided to each individual authorized to operate the pager in the paging system 10, which module is a pager. Can be movably connected to. Each codebook module can preferably be created in a facility that can be used for "education" of a pager (or a simulated equivalent like it). In this way, the burden of retraining each pager is avoided, so that each subscriber has its own codebook with them.

Each codebook module (or card) includes enough memory to store codebooks for at least one individual. Examples of memory memory cards suitable for use as codebook cards include DS1217 manufactured by Dallas Semiconductor Corporation or a functional equivalent thereof. In addition, the codebook module may include subscriber identification code, access priority code, group association code, or other code that allows subscriber access to advanced system features.

Referring again to FIG. 2, before the subscriber uses the pager 14 ', his (her) codebook module 30 must be inserted into any suitable repository within the pager. Optionally, the codebook module can be used as a "key" for the pager, such that no unauthorized person can never receive a message that can be virtually exclusive without the codebook module. Although the non- delegated person may be able to reason for the subscriber codebook module, the pager is fully functional because the pager does not respond to any voice commands other than the delegated subscriber (due to speaker dependence on good speech recognition technology). You will not.

After the codebook card is inserted into the subscriber unit 10, the subscriber controls the subscriber's partly (or wholly) by voice command. Instead, voice commands can be used in conjunction with manually entered commands (or vice versa) to perform the desired function. Whenever an individual wants to use the subscriber unit (e.g. in case of job movement changes), the codebook card can be easily removed and replaced with a new subscriber's codebook card.

Referring to FIG. 3, a block diagram of phaser 14 or 14 'is shown. The pager operates under the control of a microprocessor 32, which reads the read only memory (ROM) 38 and the random access memory (RAM) 40 via the address bus 34 and the data bus 36. To communicate with The codebook for each individual resides in a memory means (preferably electronic erase and programmable read only memory (EEPROM)) inside the codebook module 30. In the first embodiment of the present invention, the EEPROM is preferably pager 14 ) Will exist inside.

In the second embodiment, the codebook module 30 is configured and arranged to be movably connected 42 to the address bus 34 and the data bus 36 via any suitable connector or connector. Also, in another image of the second embodiment of the present invention, the microprocessor 32 operates to detect the presence of the codebook card 30, after which the microprocessor may allow the pager 14 'to operate.

Preferably, phaser 14 is controlled entirely by voice commands received and digitalized via microphone 48 prior to being processed by microprocessor 32. Optionally, the subscriber can enter a manual command via a control key (button or switch) 44. If a voice command is only partially used to control pager 14, a manually entered command 44 may be used in connection with one or more voice commands for full control of the pager.

Paging subscriber 14 receives data or voice messages via antenna 52 and receiver 54, which receive data messages to microprocessor 32 for processing, storage and routing to display 46. While proceeding, route the alert tone and voice message to any suitable audio processing stage 56, which may provide the subscriber with the tone and voice message via speaker 58. It will be appreciated by those skilled in the art that the RAM 40 includes multiple storage areas for storing certain data messages. As will be appreciated, the data message can be retrieved, deleted, protected, or processed by the subscriber by issuing an appropriate voice command. The microprocessor 32 controls 66 the battery saving circuitry 68, which controls 70 switching on / off of the receiver 54 and thus the power supply (battery) of the pager. Extends life

Optionally, the paging subscriber may include a transmitter 60, which may send an acknowledgment message from the paging subscriber 14 to the paging transmitter. Of course, if phaser 14 includes an optional transmitter 60, an antenna switch 62 and its associated antenna control line 64 may be required.

In accordance with the present invention, significant level voice control can be achieved by a limited set of words. While overall voice control requires some additional words, the present invention contemplates that pager 14 or 14 'can be at least partially controlled by the word set shown in Table 1 below.

[Figure 1]

According to the invention, the phaser 14 can be sufficiently controlled by voice commands. Voice commands can be used to activate and control advanced paging features without requiring additional manual control means such as switches or buttons. By voting one or more of the above examples of command words, the present invention can expect several advanced paging features, some of which are illustrated in Table 2 below.

[Table 2]

Some functions of these commands are, for example, "display message feed" (pager response by display of information including message five on message display and message five on message display); "Cancel message to" (pager response by deleting information including message to message from message storage means); Or "next message" (a pager response by the next message display regardless of the message number). Other features include, for example, "volume 3" (pager response by increasing or decreasing the volume to level 3) "protected message five" (pager response by protecting the message from being lost, even by a "message before deletion" command). More sophisticated as Of course, other features can be easily added by simply adding the appropriate command word to the codebook.

TECHNICAL FIELD The present invention relates to speech recognition, and more particularly, to speaker dependent speech recognition applied to a paging communication system, and a method for recognizing and responding to voice commands to program pagers.

1 is a diagram illustrating a paging system according to a first embodiment of the present invention.

2 is a view showing a pager according to a second embodiment of the present invention.

3 is a block diagram of the pager shown in FIG. 1 or FIG.

Claims (6)

In a paging system with at least one paging receiver at least partially controlled by voice commands, and at least one repository station having a database with speech recognition information for several people, a particular individual CLAIMS 1. A method of programming at least one pager to recognize the voice command of C, comprising: receiving radio frequency communication from the repository with data representing voice recognition information for the particular individual, and the voice In response to a command, processing an audio frequency voice command and at least a portion of the received data indicative of the speech recognition information to at least partially control the operation of the pager. . A pager 14, 14 ′, which is at least partially controlled and operated via voice commands, from one or more functions in the pager 14, 14 ′, from a reservoir 12 with voice recognition information for multiple individuals. Radio receiving means (52, 54, 62) for receiving speech recognition data of each individual indicative of a voice command required for initiating a voice message; storage means (30, 40) for storing the speech recognition data; Audio receiving means (48,50) for receiving a voice signal to provide a voice signal, and coupled to said storage means (30,40), at least a portion of said stored voice recognition data to recognize a voice command and said Processing means (32) for processing the received voice signal, and means (38) for initiating a function in the pager designated by the voice command in response to the voice command. A paging communication system, comprising: (a) a database of voice data for a plurality of individuals, and (b) a phone type communication in which voice recognition data identifies any one of a plurality of individuals present in the database. at least one reservoir comprising telephone interface means for receiving communication, and (c) radio transmission means for transmitting the identified individual's voice recognition data to at least one pager, and (d) outside the pager; Audio receiving means for receiving an audible speech signal generated in the system, (e) radio receiving means for receiving speech recognition data from the storage, and (f) storing speech recognition data received from the storage. And (g) at least a portion of speech recognition data tailored to said memory means to recognize a voice command and said received means. It means for processing satellite signals, and (h) paging communication system comprising the at least one pager and means for processing said voice commands to control the operation of the pager, at least in part. The voice command of claim 1, wherein at least a portion of data indicative of the speech recognition information to at least partially control the operation of the pager in response to the voice command when no manually entered command is detected. And when the manually entered command is detected, in response to the manually entered command and the voice command, processed with data representing the speech recognition information to at least partially control the operation of the pager. And in a pager to process the manually entered command in cooperation with the voice command. 3. The pager according to claim 2, wherein said processing means 32 processes said pager in response to said manually entered command by processing means for detecting the absence of a voice command and a manually entered command in the absence of said voice command. And at least partially controlling means. A paging communication method, comprising: receiving telephone type communication identifying at least one of a plurality of individuals for which speech recognition data exists in a database containing speech recognition data for the plurality of individuals, the identification In at least one repository for transmitting speech recognition data for the intended individual to at least one pager, receiving the speech recognition data from the repository, storing the speech recognition data received from the repository, the pager Receive an externally generated audible voice signal, process at least a portion of the voice recognition data stored in the memory means and the received voice signal to recognize a voice command, and at least partially control the operation of the pager At least one to process the voice command Paging communication method comprising the steps in the pager.