Abstract: In a method for efficiently and accurately measuring the intelligibility of speech, a user may utter a sample text, and an automatic speech assessment (ASA) system may receive an acoustic signal encoding the utterance. An automatic speech recognition (ASR) module may generate an N-best output corresponding to the utterance and generate an intelligibility score representing the intelligibility of the utterance based on the N-best output and the sample text. Generating the intelligibility score may involve (1) calculating conditional intelligibility value(s) for the N recognition result(s), and (2) determining the intelligibility score based on the conditional intelligibility value of the most intelligible recognition result. Optionally, the process of generating the intelligibility score may involve adjusting the intelligibility score to account for environmental information (e.g., a pronunciation score for the user's speech and/or a confidence score assigned to the 1-best recognition result).

Abstract: A method for creating personalized lesson recommendations for a user is provided. In some embodiments, the method includes assigning a skill profile to the user, where the skill profile includes a skill vector associated with one or more fine-grained skills of the user. The method also includes assigning a difficulty profile to a task to be practiced by the user, where the difficulty profile includes a task difficulty vector associated with one or more fine-grained skills of the task. Further, the method include prioritizing the task within a list of recommended tasks offered to the user based on a comparison between the skill profile of the user and the difficulty profile of the task.