Google Inc.

Pearson Education, Inc.

Jian Cheng

Peter Foltz

Lakshmi Ramachandran

In a method for quantitatively evaluating a pronunciation of a speaker, an acoustic signal is received from the speaker, which represents an utterance spoken in a language by the speaker. The acoustic signal is segmented into segments at the phone, word or phrase level. The acoustic signal segments are each transcribed by a high fidelity transcriber into high fidelity transcription segments that each comprises a sequence of phones that represents how the native or normative speakers of the language would likely perceive a playback of the acoustic signal segment. Each of the high fidelity transcription segments are compared to a baseline to generate one or more pronunciation scores that evaluate the pronunciation of the speaker. The baseline may be generated by transcribing the acoustic signal into an adapted transcription that comprises a sequence of phones that represents how native or normative speakers of the language would likely speak the utterance.

Methods for evaluating the pronunciation of speech

Systems and methods for alerting a third party device according to a result of automated evaluation of a received response packet according to at least one pattern in the data packet are disclosed herein. The system can include a memory having a pattern database and a model database. The system can further include a user device and a content management server. The content management server can receive a response packet and identify one or several patterns within the response packet. Based on the presence or absence of these one or several patterns in the response packet, the content management server can generate an evaluation of the response packet and generate and send an alert with the results of the evaluation.

Growing community of inventors

Palo Alto, CA, United States of America

Jian Cheng