System and method for correcting accent induced speech transmission problems

摘要

A system and method is provided for detecting errors in a speech transmission system. A first audio stream is comprised of a plurality of words, upon which a plurality of independent voice-to-text conversions are performed. If it is determined that at least one of the plurality of independent voice-to-text conversions is error free, a text-to-voice conversion of the at least one error-free voice-to-text conversion is performed to create a second audio stream.

主权项

1. A method for detecting errors in a speech transmission system, the method comprising:
receiving a first audio stream comprised of a plurality of words having an order; generating digitized audio by recording the first audio stream; performing at least two independent voice-to-text conversions of the audio stream thereby generating a first and a second text version of the plurality of words; comparing each word of the plurality of words in the first and second text versions, respectively, thereby generating (i) a first set of words comprising words for which the first and second text versions are identical and (ii) a second set of words comprising words for which the first and second text versions differ; performing a text-to-voice conversion of the first set of words; identifying parts of the digitized audio corresponding to the words of the second set of words; and creating a second audio stream comprising (i) the text-to-voice converted first set of words and (ii) the parts of the digitized audio corresponding to the words of the second set of words, such that words in the second audio stream have the same order as the plurality of words in the first audio stream.