Computer-Implemented Systems and Methods for Determining an Intelligibility Score for Speech

摘要

Systems and methods are provided for generating an intelligibility score for speech of a non-native speaker. Words in a speech recording are identified using an automated speech recognizer, where the automated speech recognizer provides a string of words identified in the speech recording, and where the automated speech recognizer further provides an acoustic model likelihood score for each word in the string of words. For a particular word in the string of words, a context metric value is determined based upon a usage of the particular word within the string of words. An acoustic score for the particular word is determined based on the acoustic model likelihood score for the particular word from the automated speech recognizer. An intelligibility score is determined for the particular word based on the acoustic score for the particular word and the context metric value for the particular word.

主权项

1. A computer-implemented method of generating an intelligibility score for speech of a non-native speaker, comprising:
receiving a recording of speech of a non-native speaker at a processing system; identifying words in the speech recording using a computerized automated speech recognizer, wherein the automated speech recognizer provides a string of words identified in the speech recording based on a computerized acoustic model, and wherein the automated speech recognizer further provides an acoustic model likelihood score for each word in the string of words; for a particular word in the string of words, determining a context metric value with the processing system based upon a usage of the particular word within the string of words; determining an acoustic score with the processing system for the particular word based on the acoustic model likelihood score for the particular word from the automated speech recognizer; determining an intelligibility score with the processing system for the particular word based on the acoustic score for the particular word and the context metric value for the particular word; and determining an overall intelligibility score with the processing system for the string of words based on the intelligibility score for the particular word and intelligibility scores for other words in the string of words.