Loading patent…
AssigneeNuance Communications, Inc.
InventorsThomas Murray Kuhn, Matthew Lennig, Peter Christopher Monaco, David Bruce Peters
Estimating the term…
A system and method for efficiently distributing voice call data received from speech recognition servers over a telephone network having a shared processing resource is disclosed. Incoming calls are received from phone lines and assigned grammar types by speech recognition servers. A request for processing the voice call data is sent to a resource manager which monitors the shared processing resource and identifies a preferred processor within the shared resource. The resource manager sends an instruction to the speech recognition server to send the voice call data to a preferred processor for processing. The preferred processor is determined by known processor efficiencies for voice call data having the assigned grammar type of the incoming voice call data and a measure of processor loads. While the system is operating, the resource manger develops and updates a history of each processor. The histories include processing efficiency values for all grammar types received. The processing efficiencies are stored, tabulated and assigned usage number values for each processor. When incoming voice call data is receive, the resource manages evaluates the total sum of the usage numbers for processing requests assigned to each processor and the usage number for the grammar type of the incoming data as applied to each processor. The incoming data is distributed to the processor with the lowest sum of total of usage numbers for assigned requests and the usage number assigned to the incoming data for that processor.
A system architecture for efficiently distributing voice call data over a telephone network using speech recognition, the system architecture comprising:
A system for efficiently distributing data over a telephone network using speech recognition, the system comprising:
A method for distributing call data over a telephone network with a plurality of processing stations, the method comprising the steps of:
An adaptive system for efficiently distributing data over a telephone network using speech recognition, the system comprising
Dependencies are inferred from the claim text.
Drawings are temporarily unavailable. Open the PDF at USPTO
No related US applications are recorded in the indexed filing history.
Constructional details of speech recognition systems: distributed recognition, e.g. in client-server systems, for mobile phones or network…
Speech recognition: procedures used during a speech recognition process, e.g. man-machine dialogue
Speech analysis techniques or speech synthesis
Term ended 13 Mar 2018 (estimated)
USPTO file history isn’t available for this patent, so only indexed dates are shown.
This invention relates to the field of voice processing. More particularly, this invention relates to an improved architecture having a distributed work allocation control system for a voice processing system.
Automated systems for recognizing spoken natural language sequences require varying amount of processing capacity dependent upon the nature of the spoken message. It is well understood that it takes relatively less processor attention to recognize a string of spoken digits than to recognize the spoken name of an individual from among a list of thousands or even hundreds of thousands.
To appreciate this artifact of speech recognition, consider the number of potential words necessary to recite the numbers from zero to 250,000. There are the ten words for single digit numbers: `one`, `two`, `three`. . . , the ten words for the teen numbers: `eleven`, `twelve`, `thirteen`, . . . , the ten words for the decades: `ten`, `twenty`, `thirty`, . . . , and the two words, for larger place identification `hundred` and `thousand`. This relatively limited list of words for speaking a numerical string can result in rather simple or efficient processing of such spoken strings. In contrast, consider a telephone directory having 250,000 names of individuals. Each such name is potentially quite different from one another and can include names such as `Smith`, `Jones`, `Yamasaki` and `Van Rysselberghe`. It is clear that the solution for recognizing such a diverse collection of audible sounds can be more difficult than for recognizing a string of numerical digits.
FIG. 1 shows a block diagram of a prior art speech recognition system.
FIG. 2 shows a block diagram of the preferred embodiment of the present invention.
FIG. 3 shows a table used in assignment of utterances for one resource manager in the system of FIG. 2.
FIG. 2 shows a block diagram of the preferred embodiment of the present invention. The system is adapted to identify and handle a variety of different types of speech calls in an efficient manner.
A telephone line 100 is coupled to provide a user access to the natural language speech recognition system 102 of the present invention. The telephone line 100 is coupled to a RecCient 104. Under certain circumstances, such as shown in FIG. 2, a plurality of RecClients 104 are coupled within the system to allow simultaneous processing of a plurality of calls received over a plurality of telephone lines 100. The RecClient 104 is adapted to perform certain speech preprocessing operations such as determining when a call begins (going off-hook), when a call ends (going on-hook), when speech begins and when speech stops.
A speech application 106 is coupled to the RecClient 104. The speech application 106 controls certain aspects of the operation of the RecClient 104. For example upon determining that a call has begun, the speech application 106 will cause the RecClient 104 to play a user prompt "Welcome to the system." If the speech application 106 detects that the user is speaking while a user prompt is playing, the speech application 106 will instruct the RecClient 104 to stop playing the user prompt. The speech application 106 and the RecClient 104 are configured to cancel echo and noise on the telephone line. These examples are merely exemplary and are ancillary to the present invention. Many other forms of control and response can be exhibited by the speech application 106 and still fall within the spirit and scope of this invention.