Apparatus and methods for rejecting confusible words during training associated with a speech recognition system

Computer-based apparatus for training at least one new word for addition to a vocabulary of a speech recognition engine containing existing words, the apparatus comprising:

• an input device for receiving the at least one new word uttered by a user;
• a processor, operatively coupled to the input device, for computing respective measures between the at least one newly uttered word and at least a portion of the existing vocabulary words, the respective measures indicative of acoustic similarity between the at least one word and the at least a portion of existing words;
• and
• if no measure is within the threshold range, the processor automatically adding the at least one newly uttered word to the vocabulary, and if at least one measure is within a threshold range, the processor refraining from automatically adding the at least one newly uttered word to the vocabulary.

Computer-based apparatus for training at least one new word for addition to a vocabulary of a speech recognition engine containing existing words, the apparatus comprising:

• user input means for receiving at least one new word uttered by the user;
• computing means for computing respective measures between an acoustic model of the at least one newly uttered word and acoustic models of at least a portion of the existing vocabulary words, the respective measures indicative of acoustic similarity between the at least one word and the at least a portion of existing words;
• adding means for automatically adding the at least one newly uttered word to the vocabulary, if no measure is within the threshold range;
• and
• rejecting means for automatically rejecting the at least one newly uttered word, if at least one measure is within a threshold range.