Recognizing the numeric language in natural spoken dialogue
View Patent ↗A system and a method are provided. A speech recognition processor receives unconstrained input speech and outputs a string of words. The speech recognition processor is based on a numeric language that represents a subset of a vocabulary. The subset includes a set of words identified as being for interpreting and understanding number strings. A numeric understanding processor contains classes of rules for converting the string of words into a sequence of digits. The speech recognition processor utilizes an acoustic model database. A validation database stores a set of valid sequences of digits. A string validation processor outputs validity information based on a comparison of a sequence of digits output by the numeric understanding processor with valid sequences of digits in the validation database.
1. A method of processing speech, the method comprising:
receiving speech and generating a string of words based on the speech and based at least in part on a numeric language including a set of words for recognizing and processing number strings;
converting the string of words into a sequence of digits using classes of rules and according to an acoustic model database in which Markov models characterize acoustic features of numeric words and each word is modeled by three segments comprising a head, a body and a tail, and wherein the three segment structure only applies to single digit words;
storing a set of valid sequences of digits; and
generating validity information based on a comparison of the sequence of digits with a valid sequences of digits of the stored set of valid sequences of digits.
2. The method of claim 1 , wherein the valid sequences of digits comprise valid telephone numbers.
3. The method of claim 1 , wherein the valid sequences of digits comprise valid credit card numbers.
4. The method of claim 1 , wherein the valid sequences of digits comprise valid social security numbers.
5. The method of claim 1 , further comprising:
identifying out-of-vocabulary utterances and utterances that are poorly recognized.
6. The method of claim 1 , further comprising:
initiating an action based on the validity information.
7. The method of claim 1 , further comprising:
storing data describing a structure and a sequence of words and phrases.
8. The method of claim 1 , wherein the classes of rules include a restarts rule.
9. A tangible computer-readable medium storing instructions for controlling a computing device to process speech, the instructions causing the computing device to perform steps comprising:
receiving speech and generating a string of words based on the speech and based at least in part on a numeric language including a set of words for recognizing and processing number strings;
converting the string of words into a sequence of digits using classes of rules and according to an acoustic model database in which Markov models characterize acoustic features of numeric words and each word is modeled by three segments comprising a head, a body and a tail, and wherein the three segment structure only applies to single digit words;
storing a set of valid sequences of digits; and
generating validity information based on a comparison of the sequence of digits with a valid sequences of digits of the stored set of valid sequences of digits.
10. The tangible computer-readable medium of claim 9 , wherein the valid sequences of digits comprise valid telephone numbers.
11. The tangible computer-readable medium of claim 9 , wherein the valid sequences of digits comprise valid credit card numbers.
12. The tangible computer-readable medium of claim 9 , wherein the valid sequences of digits comprise valid social security numbers.
13. The tangible computer-readable medium of claim 9 , the instructions further comprising:
identifying out-of-vocabulary utterances and utterances that are poorly recognized.
14. The tangible computer-readable medium of claim 9 , the instructions further comprising:
initiating an action based on the validity information.
15. The tangible computer-readable medium of claim 9 , the instructions further comprising:
storing data describing a structure and a sequence of words and phrases.
16. The tangible computer-readable medium of claim 9 , wherein the classes of rules include a restarts rule.
17. A system for processing speech, the system comprising:
a processor;
a module that controls the processor to generate a string of words based on received speech and based at least in part on a numeric language including a set of words for recognizing and processing number strings;
a module that controls the processor to convert the string of words into a sequence of digits using classes of rules and according to an acoustic model database in which Markov models characterize acoustic features of numeric words and each word is modeled by three segments comprising a head, a body and a tail, and wherein the three segment structure only applies to single digit words;
a module that controls the processor to store a set of valid sequences of digits; and
a module that controls the processor to generate validity information based on a comparison of the sequence of digits with a valid sequences of digits of the stored set of valid sequences of digits.
18. The system of claim 17 , wherein the valid sequences of digits comprise valid telephone numbers, credit card numbers and social security numbers.
19. The system of claim 17 , further comprising:
a module that controls the processor to initiate an action based on the validity information.
20. The system of claim 17 , further comprising:
a module that controls the processor to store data describing a structure and a sequence of words and phrases.