Telephonic communications – Audio message storage – retrieval – or synthesis – Voice activation or recognition
Reexamination Certificate
1998-12-31
2004-06-01
Tsang, Fan (Department: 2645)
Telephonic communications
Audio message storage, retrieval, or synthesis
Voice activation or recognition
C379S088010, C379S207020, C455S563000
Reexamination Certificate
active
06744860
ABSTRACT:
FIELD OF THE INVENTION
The present invention is directed to communications systems and, more particularly, to communications systems which support voice operations, e.g., voice-dialing operations.
BACKGROUND OF THE INVENTION
People frequently have difficulty remembering all of the telephone numbers that they use. In addition, many people, e.g., physically handicapped individuals, find it difficult to manually enter a telephone number into a telephone set, e.g., by depressing buttons on a telephone.
In order to facilitate dialing of telephone numbers, voice-dialing telephone services have been introduced. Once connected to a service provider, known voice-dialing services allow a caller to dial a telephone number by simply speaking a dial command followed by the name of the party to be called and/or by simply stating the name of the party to be called. Speech recognition is used in the known voice-dialing services to identify the party to be called. Once the spoken name of the party is identified, the voice-dialing service dials a telephone number associated with the identified name, thereby eliminating the need for the voice-dialing subscriber to enter the telephone number of the party to be called.
Known voice-dialing services of the type described above are frequently implemented using an intelligent peripheral, e.g., a voice-dialing platform with speech recognition capabilities, coupled to a central office (CO) switch.
FIG. 1
illustrates a representative known telephone system
10
. As illustrated, the telephone system
10
comprises first and second telephones
12
, cell phone
14
with antenna
15
, and a public switched telephone network (PSTN)
16
. PSTN
16
comprises a CO switch
20
, a mobile telephone switching office (MTSO) switch
22
and an intelligent peripheral (IP)
18
capable of performing voice-dialing services. The MTSO switch
22
includes an antenna
23
for communicating with cell phone
14
via antenna
15
. The first and second telephones
12
may be, e.g., MINDSET telephones produced by XINEX corporation which include an RS
232
interface.
One problem with the system
10
is that a call must be connected to the IP
18
before a voice-dialing service can be provided. In the known system
10
, the caller may contact the IP
18
in two ways: the first way involves dialing a telephone number corresponding to the IP
18
; and the second way involves initiating an off-hook condition, which automatically results in connecting a voice-dialing subscriber to the IP
18
for voice-dialing services. The second approach for connecting a voice-dialing subscriber involves the use of what has become known as the NFA protocol. A description of the second approach for connecting a service subscriber to an IP can be found in U.S. Pat. No. 5,784,437.
Use of the NFA protocol can eliminate the need to perform a dialing operation in order to contact an IP
18
so that a voice-dialing service can be performed. As such, use of the NFA protocol eliminates the need to remember and enter the telephone number of the IP
18
that provides a voice-dialing service.
Unfortunately, even with the use of the NFA protocol, an off-hook condition must be initiated before a voice-dialing operation can be performed. The physical action of lifting a handset or pressing a button to initiate an off-hook condition can be difficult or impossible for some handicapped individuals. In addition, from a convenience standpoint, it is desirable that the need to perform a physical lifting or switching act be eliminated so that a telephone call can be initiated without the use of hands. The ability to perform a hands-free dialing operation is appealing, from a convenience standpoint, to many business, home and cellular phone users. In addition for cellular phone users operating a motor vehicle, the ability to initiate a dialing operation without having to use one's hands offers important safety advantages to methods which require the removal of a hand from a steering wheel.
SUMMARY OF THE PRESENT INVENTION
The present invention is directed to communications systems and methods and, more particularly, to methods and apparatus for implementing hands-free dialing operations.
In accordance with one embodiment of the present invention, an intelligent speakerphone, when not in an off-hook condition, is programmed to monitor ambient sounds to detect particular speech corresponding to a command to initiate a voice-dialing operation. The speakerphone's microphone may be used to detect audio signals, which are processed by a processing unit within the speakerphone to detect a voice-dialing command. When off-hook the speakerphone may monitor for a verbal hang-up command.
Upon detecting a command or other acoustic signal used to initiate a voice-dialing operation, the speakerphone of the present invention initiates an off-hook condition. In an embodiment where the NFA protocol is used, the off-hook condition results in the caller being connected to a voice-dialing IP, thereby allowing the user of the telephone to proceed with a voice-dialing call. In embodiments where the NFA protocol is not supported, e.g., in some cellular telephone embodiments, the speakerphone of the present invention automatically dials the voice-dialing IP after initiating an off-hook condition. The dialing is automatically performed in response to the same command or signal used to initiate the voice-dialing operation. The telephone of the present invention may automatically provide the voice-dialing IP a subscriber's personal identification number (PIN) in embodiments where such information is used to perform a voice-dialing operation.
Once connected to the IP, the IP performs a voice dialing operation in a conventional manner. In such an embodiment, the IP performs speech recognition of names of individuals and outputs the telephone number's associated therewith. In this manner, the IP which is a shared resource is responsible for the majority of the speech recognition required to perform a complete voice dialing operation which the telephony device of the present invention only has to be capable of recognizing a single word or phrase corresponding to a dial command. In such an embodiment, the IP will normally support a vocabulary that is many times larger than the vocabulary supported by the telephony device used to initiate the voice dialing operation.
Since many speakerphones currently include processors, memory devices, and microphones suitable for implementing the present invention, the voice-dialing feature of the present invention may be implemented at little or no expense, in terms of hardware, beyond that already associated with manufacturing many modern speakerphones.
In addition to a speakerphone embodiment, the voice-dialing initiation method of the present invention may be used with computers capable of performing telephony operations, telephones capable of being controlled by an external device, e.g., a voice-dialing control circuit of the present invention, cellular telephones and a wide variety of other telephony devices. While the methods and apparatus of the present invention are described in the context of a voice-dialing embodiment, it is to be understood that the speech recognition and monitoring techniques of the present invention may be used to initiate other services, e.g., voice mail services, requiring control of a communications device and/or connection to an intelligent peripheral.
Speech recognition models used by the telephony devices of the present invention are, in one embodiment, generated by an IP from speech forwarded to the IP from the telephony devices. The speech recognition models may be speaker dependent models generated from an individual speaker and used by an individual telephony device or speaker independent models generated from multiple speakers. Speech recognition models generated by the IP are stored in and used by the individual telephony devices to detect signals or commands to initiate a voice dialing operation. By placing the burden of speech recognition model generati
Bell Atlantic Network Services
Escalante Ovidio
Straub Michael P.
Straub & Pokotylo
Tsang Fan
LandOfFree
Methods and apparatus for initiating a voice-dialing operation does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Methods and apparatus for initiating a voice-dialing operation, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Methods and apparatus for initiating a voice-dialing operation will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3303514