Recognizing the numeric language in natural spoken dialogue
View Patent ↗A system for recognizing connected digits in natural spoken dialogue includes a speech recognition processor that receives unconstrained fluent input speech and produces a string of words that can include a numeric language, and a numeric understanding processor that converts the string of words into a sequence of digits based on a set of rules. An acoustic model database utilized by the speech recognition processor includes a first set of hidden Markov models that characterize the acoustic features of numeric words and phrases, a second set of hidden Markov models that characterize the acoustic features of the remaining vocabulary words, and a filler model that characterizes the acoustic features of out-of-vocabulary utterances. An utterance verification processor verifies the accuracy of the string of words. A validation database stores a grammar, and a string validation processor outputs validity information based on a comparison of the sequence of digits with the grammar. A dialogue manager processor initiates an action based on the validity information.
1. A speech recognition method, comprising:
receiving a speech signal;
performing a speech recognition process on the received speech signal to produce speech recognition results, the speech recognition process being based on a set of acoustical models that has been defined for a numeric language, wherein the numeric language includes a subset of a vocabulary, the subset of the vocabulary including words that identify digits in number strings and words that enable the interpretation and understanding of number strings; and
generating a sequence of digits using said speech recognition results, said generating being based on a set of rules, wherein
the numeric language includes digits, natural numbers, alphabets, re-starts, and city/country name classes.
2. The method of claim 1 , wherein the acoustical models are hidden Markov models.
3. The method of claim 1 , wherein said generating is implemented by a numeric understanding processor.
4. The method of claim 1 , wherein the set of rules includes one of a naturals rule and alphabets rule.
5. The method of claim 1 , wherein the set of rules includes a restarts rule.
6. The method of claim 1 , wherein the set of rules includes a city/country rule.
7. The method of claim 1 , wherein the set of rules includes a numeric phrases rule that realigns digits.
8. The method of claim 1 , wherein said performing is implemented by a speech recognition processor.
9. The method of claim 1 , wherein said performing is further based on a second set of acoustical models that has been defined for other words in the vocabulary.
10. The method of claim 9 , wherein said second set of acoustical models is defined at a quality level different than the set of acoustical models for the numeric language.
11. A system comprising:
a speech recognition processor that receives unconstrained input speech and outputs a string of words, the speech recognition processor being based on a numeric language that represents a subset of a vocabulary, the subset including a set of words identified as being relevant for interpreting and understanding number strings;
a numeric understanding processor containing classes of rules for converting the string of words into a sequence of digits;
an acoustic model database utilized by the speech recognition processor;
a validation database that stores a set of valid numbers; and
a string validation processor that outputs validity information based on a comparison of a sequence of digits output by the numeric understanding processor with valid numbers in the validation database, wherein
said numeric language includes digits, natural numbers, alphabets, re-starts, and city/country name phrase classes.