Automated recognition system for natural language understanding
View Patent ↗An interactive response system directs input to a software-based router, which is able to intelligently respond to the input by drawing on a combination of human agents, advanced recognition and expert systems. The system utilizes human “intent analysts” for purposes of interpreting customer input. Automated recognition subsystems are trained by coupling customer input with IA-selected intent corresponding to the input, using model-updating subsystems to develop the training information for the automated recognition subsystems.
1. A system for processing an interaction with a person, comprising:
a routing processor configured to receive data representing an input provided by the person;
an analyst user interface device in communication with the routing processor, configured to present to at least one human analyst information related to the input in perceptible form and to accept, during an interaction between the person and an interactive response system, a business meaning associated with the intent from the at least one human analyst; and
a training subsystem configured to:
receive the information and the business meaning;
select a target vocabulary based on the business meaning, the target vocabulary constraining interpretation of input to a given set of words;
provide machine learning services to a non-real time training recognizer based on the selected target vocabulary and the input, the non-real time training recognizer comprising at least one of an automated speech recognizer (ASR), machine vision, or natural language processor
update, by the non-real time training recognizer using the selected target vocabulary and the input, a mapping between the business meaning and one or more words in the target vocabulary, the mapping including statistics indicating, for each of the one or more words, a probability of input from a person representing that word; and
train, via the statistics, a model used by a real-time recognizer.
2. The system of claim 1 , the system further comprising a real-time recognizer in communication with the routing processor and configured to receive the data representing the input, wherein the training subsystem is configured to train the real-time recognizer by updating the model.
3. The system of claim 1 , the system further comprising a real-time recognizer in communication with the routing processor and configured to receive the data representing the input, wherein the training subsystem is configured to continue training the real-time recognizer responsive to the real-time recognizer not reaching a performance threshold.
4. The system of claim 1 , wherein the training subsystem is further configured to:
determine a level of end-user traffic; and
based on the level of end-user traffic, determine a number of intent analysts to assign to interpret the input.
5. The system of claim 1 , wherein the input is speech.
6. The system of claim 1 , wherein the input is text.
7. A computer-implemented method for operating an interactive response system comprising:
receiving data representing an input from a person;
using a processor, automatically presenting information relating to the input to at least one intent analyst through an analyst user interface;
receiving, during an interaction between the person and the interactive response system, a business meaning associated with the input from the at least one intent analyst, the business meaning provided by the at least one intent analyst through the analyst user interface;
selecting, by a training subsystem, a target vocabulary based on the business meaning, the target vocabulary constraining interpretation of input to a given set of words;
providing, by the training subsystem, machine learning services to a non-real time training recognizer based on the selected target vocabulary and the input, the non-real time training recognizer comprising at least one of an automated speech recognizer (ASR), machine vision, or natural language processor
updating, by the non-real time training recognizer using the selected target vocabulary and the input, a mapping between the business meaning and one or more words in the target vocabulary, the mapping including statistics indicating, for each of the one or more words, a probability of input from a person representing that word; and
training, via the statistics, a model used by a real-time recognizer in order to improve performance thereof.
8. The computer-implemented method of claim 7 , further comprising:
determining a level of end-user traffic; and
based on the level of end-user traffic, determining a number of intent analysts to assign to interpret the input.
9. The computer-implemented method of claim 7 , wherein the training comprises updating the model.
10. The computer-implemented method of claim 7 , wherein the training comprises testing performance of the real-time recognizer and continuing training responsive to the performance not exceeding a performance threshold.
11. The method of claim 7 , wherein the input is speech.
12. The method of claim 7 , wherein the input is text.
13. A system for managing interactions with a person, comprising non-transitory computer storage media storing programming instructions executable by at least one processor for:
receiving data representing an input from a person;
using a processor, automatically presenting information relating to the input to at least one intent analyst through an analyst user interface;
receiving, during an interaction between the person and the interactive response system, a business meaning associated with the input from the at least one intent analyst, the business meaning provided by the at least one intent analyst through the analyst user interface;
selecting, by a training subsystem, a target vocabulary based on the business meaning, the target vocabulary constraining interpretation of input to a given set of words;
providing, by the training subsystem, machine learning services to a non-real time training recognizer based on the selected target vocabulary and the input, the non-real time training recognizer comprising at least one of an automated speech recognizer (ASR), machine vision, or natural language processor
updating, by the non-real time training recognizer using the selected target vocabulary and the input, a mapping between the business meaning and one or more words in the target vocabulary, the mapping including statistics indicating, for each of the one or more words, a probability of input from a person representing that word; and
training, via the statistics, a model used by a real-time recognizer in order to improve performance thereof.
14. The system of claim 13 , further comprising instructions for:
determining a level of end-user traffic; and
based on the level of end-user traffic, determining a number of intent analysts to assign to interpret the input.
15. The system of claim 13 , wherein the training comprises updating the model.
16. The system of claim 13 , wherein the training comprises testing performance of the real-time recognizer and continuing training responsive to the performance not exceeding a performance threshold.
17. The system of claim 13 , wherein the input is speech.
18. The system of claim 13 , wherein the input is text.