Speaker-identification-assisted speech processing systems and methods

摘要

Methods, systems, and apparatuses are described for performing speaker-identification-assisted speech processing. In accordance with certain embodiments, a communication device includes speaker identification (SID) logic that is configured to identify a user of the communication device and/or the identity of a far-end speaker participating in a voice call with a user of the communication device. Knowledge of the identity of the user and/or far-end speaker is then used to improve the performance of one or more speech processing algorithms implemented on the communication device.

主权项

1. A communication device, comprising:
speaker identification logic configured to apply a speaker identification algorithm to a speech signal to generate speaker identification information, the speaker identification information including at least an identifier that identifies a target speaker associated with the speech signal; and speech processing logic comprising a plurality of speech signal processing stages, wherein each of the plurality of speech signal processing stages is configured to process the speech signal in accordance with a respective speech processing algorithm based on the speaker identification information provided by the speaker identification logic, wherein the speaker identification logic is further configured to apply the speaker identification algorithm to the speech signal to generate a first measure of confidence that is indicative of the likelihood that the speech signal is associated with a target speaker; wherein a first speech signal processing stage of the plurality of speech signal processing stages is configured to process the speech signal in accordance with a first speech processing algorithm in a manner that takes into account the first measure of confidence to produce a processed speech signal; and wherein the speaker identification logic is further configured to apply the speaker identification algorithm to the processed speech signal to generate a second measure of confidence that is indicative of the likelihood that the processed speech signal is associated with the target speaker.