Telephonic communications – Audio message storage – retrieval – or synthesis – Message signal analysis
Reexamination Certificate
1998-02-05
2002-02-12
Weaver, Scott L. (Department: 2645)
Telephonic communications
Audio message storage, retrieval, or synthesis
Message signal analysis
C379S386000
Reexamination Certificate
active
06347134
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention pertains to improvements in the field of touch tone telephone systems having voicemail recording capability. Specifically, the invention provides the ability to prevent voices having DTMF mimic qualities from causing false DTMF commands to the system from being issued during the process of recording voicemail messages.
Touch tone telephone systems generate DTMF tones when keys on the phone are pressed for such purposes as dialing telephone numbers, giving commands to voicemail systems etc. DTMF tones, generally, are different combinations of audio frequencies generated simultaneously. For example, when the three key is pressed to delete a voicemail message in a telephone system having voicemail messaging capability, typically two tones of different, constant frequency will be generated. The filtering and recognition circuitry in the voicemail command recognition circuitry will detect the two tones and interpret them as a three or the command to delete a voicemail message either just played or in the process of being played.
A problem arises because of the need in touch tone telephone systems with voicemail messaging capability for voice and DTMF tones to be sent on the same communication channel. Human voices generate complex combinations of tones frequently. Certain human voices have the property that at random times combinations of tones are generated which are the same as DTMF tones. Such instances are called DTMF mimics. This is an unfortunate byproduct of the fact that the original designers of touch tone telephones selected DTMF tones to be combinations of tones in the audible range where they must coexist with the audible tones generated by human voices. Now these original selections for DTMF tones have become an industry standard. Thus, designers of telephone systems for use in businesses and homes must include standard DTMF tone generators such that the DTMF tones they generate will be properly interpreted by switching equipment in the central office of the telephone company—equipment over which the designer of the telephone system connected to the central office has no control.
Had the original designers of touch tone telephone systems selected DTMF tones that were outside the range of tones generated by human voices, the problem would not exist. However, the selection of DTMF tones that were audible does have the advantage that it provides audible feedback to the user through the sidetone heard by the user indicating to that user that the telephone DTMF tone generator is working and that dialing or other commands have been sent.
The specific problem to which the system of the invention provides a solution is the fact that during voicemail message playback DTMF mimics can sometimes occur. When they do occur, results can be unpredictable. For example, a DTMF mimic that commanded “Repeat the Voicemail Message” could cause a voicemail message to be repeated from the beginning each time the point in the message is reached where the DTMF mimic occurs. This would prevent the user from ever hearing the portion of the message past the DTMF mimic. Likewise, if the DTMF mimic is a “Delete the Voicemail Message” command, the message containing the mimic would be deleted as soon as the point in the message where the DTMF mimic occurred.
The prior art teaches systems for detecting DTMF mimics but does not teach the solution provided by the invention for improving the operation of voicemail system by eliminating the unpredictable effects of DTMI: mimics that occur during message playback.
U.S. Pat. No. 4,124,773 entitled “Audio Storage and Distribution System” to Elkins teaches the use of DTMF signalling to retrieve stored audio signals such as voicemail. It does not address the problem of DTMF mimics in playback or detecting and eliminating DTMF mimics during message recording.
DTMF mimics have been generally a problem in telephone equipment for many years. Before voicemail, the problem of DTMF mimics was handled in standard telephone equipment by only detecting DTMF tones during certain intervals. For example, prior art systems might listen for DTMF tones when a user is placing a call only until enough digits have been entered to make up a complete telephone numbers.
U.S. Pat. No. 5,528,663 filed Jul. 26, 1995 by Locke, et al., and assigned to Sieman-Robin Communications teaches a voice store and forward system with improved DTMF signal detection. This patent is addressed to the problem of “talk-off” which is the term for DTMF detection circuitry accepting a voice signal as a DTMF command thereby creating an error. Other problems addressed by this patent include distortion of DTMF signals when outgoing voice signals mix with inbound DTMF tones in the transhybrid circuit. Because of finite loss in the transhybrid, the DTMF tones will contain additive voice noise thereby interfering with reception of the DTMF signals. The patent notes that one solution to the talk-off problem proposed in the prior art was to make the minimum time for a DTMF tone longer. While this reduces the problem, it does not eliminate it and causes other problems in that some DTMF tone generators have a maximum tone generation time which is shorter than the longer standard. Another solution to talk-off proposed in the prior art was to require a certain minimum power at harmonic frequencies of the DTMF tones before accepting an alleged DTMF command as legitimate. This too did not eliminate the problem. Another solution to talk-off proposed in the prior art was to narrow the bandwidth of the individual DTMF receiver filters. This too did not completely eliminate the problem and can run afoul of the EIA specification for the minimum bandwidth of DTMF filters of 1.5% of the center frequency to account for individual variations in DTMF tone generators. The solution to talk-off proposed by the '663 Patent involves recognition of DTMF tones using a minimum of processing both power and time. The receiver taught in this prior art reference functions to both identify the DTMF tones as well as provide measurements of the power of its individual components. The receiver outputs signal power, component power and DTMF value. These outputs are provided to a rating algorithm which evaluates the quality of the input signal to determine if it is a legitimate DTMF signal. The algorithm outputs a rating value from 1 to 4 with 4 being the lowest and considered to be non-DTMF. If the algorithm outputs a rating between 1 and 3, the DTMF identification, rating and power are passed to timing/stability algorithm. This algorithm evaluates the stability of the DTMF signal. The algorithm makes its stability determination by evaluating whether the DTMF value or rating or power over time. If any one of these parameters changes over the interval in question, the signal is rejected as not DTMF.
U.S. Pat. No. 5,519,764 filed Mar. 8, 1993 by Pierce, et al., teaches a voice messaging system which scans messages for phrases which might be interpreted as DTMF tones. A record is kept by the computer as to where, when and for how long the DTMF mimic may be occurring. This allows prerecorded or synthesized Voice Response Announcement System messages to be checked to determine if an error caused by a DTMF mimic is about to occur before the message is played. The '764 patent is employed in telephone system control office systems with Voice Response Announcement Systems. In these systems, a user sends DTMF commands to the central office using his telephone. Sometimes these commands call for messages from the Voice Response Announcement System to be played such as in a voicemail user interface system. A problem arises with the message played back to the user through the central office contains DTMF mimics. If such mimics are present in the outgoing message from the central office, the DTMF receivers in the central office may react by carrying out an undesired action. The '764 Patent system screens these Voice Response Announcement System messages from DTMF mimics and catalogues them.
U.S. Pat. No
Lasswell Charles Arnold
Sherwood Charles Gregory
Picazo Communications
Weaver Scott L.
LandOfFree
Recording and playing voicemail with improved DTMF detection does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Recording and playing voicemail with improved DTMF detection, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Recording and playing voicemail with improved DTMF detection will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2948892