Telephone network service for converting speech to touch-tones
First Claim
1. A method for providing a telecommunications network service performed within a telecommunications network for use with an automated system responsive to Dual-Tone Multi-Frequency (DTMF) signals (touch-tones), the method comprising the steps of:
- prompting the caller to initiate a first call to the automated system initiating a second call in the network to a facility for performing speech recognition;
bridging the first and second calls;
receiving the spoken utterance comprising the spoken word;
performing speech recognition on the spoken utterance, thereby identifying the spoken word;
classifying the spoken utterance into one of a set of categories based on said identification of the spoken word, each of said categories having a DTMF signal associated therewith;
generating, within the telecommunications network, the DTMF signal associated with the category into which said spoken utterance has been classified; and
transmitting the generated DTMF signal though the telecommunications network to the automated system.
4 Assignments
0 Petitions
Accused Products
Abstract
A telecommunications network service for converting spoken words to individual DTMF (e.g., touch-tone) signals to be furnished to an automated system responsive to touch-tone control thereof. In accordance with one illustrative embodiment of the invention, a telephone user speaks an utterance comprising a word representing a digit, letter, the “*” key (e.g., the word “star”) or the “#” key (e.g., the word “pound”). The illustrative network service performs conventional, speaker independent speech recognition on the utterance to identify the spoken word and then generates the touch-tone signal which corresponds to the spoken word so identified (e.g., a digit, letter or special character as found on a conventional touch-tone telephone). This generated DTMF signal is then transmitted to an automated system such as a voice messaging system or telephone answering machine to control the operation thereof. The network service of the present invention may be advantageously employed when the telephone being used by the caller is a rotary or dial pulse type telephone, or in other circumstances where it may be preferable to speak rather than to press the keys of a telephone keypad.
54 Citations
9 Claims
-
1. A method for providing a telecommunications network service performed within a telecommunications network for use with an automated system responsive to Dual-Tone Multi-Frequency (DTMF) signals (touch-tones), the method comprising the steps of:
-
prompting the caller to initiate a first call to the automated system initiating a second call in the network to a facility for performing speech recognition;
bridging the first and second calls;
receiving the spoken utterance comprising the spoken word;
performing speech recognition on the spoken utterance, thereby identifying the spoken word;
classifying the spoken utterance into one of a set of categories based on said identification of the spoken word, each of said categories having a DTMF signal associated therewith;
generating, within the telecommunications network, the DTMF signal associated with the category into which said spoken utterance has been classified; and
transmitting the generated DTMF signal though the telecommunications network to the automated system. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9)
-
Specification
-
- Resources
-
Current AssigneeNuance Communications, Inc. (Microsoft Corporation)
-
Original AssigneeAT&T Corporation (AT&T, Inc.)
-
InventorsBulfer, Andrew Frederick
-
Primary Examiner(s)Tsang, Fan
-
Assistant Examiner(s)Opsasnick, Michael N.
-
Application NumberUS08/867,773Time in Patent Office1,394 DaysField of Search395/26, 379/201, 379/67US Class Current704/251CPC Class CodesH04M 1/271 controlled by voice recogni...H04M 2201/40 using speech recognition sp...H04M 2201/60 Medium conversionH04M 3/42204 Arrangements at the exchang...H04M 3/493 Interactive information ser...H04M 3/533 Voice mail systemsH04Q 1/45 using multi-frequency signa...
