Systems and methods for classification of audio environments

摘要

Systems and methods for classification of audio environments are disclosed. In one embodiment, a method of classifying an audio environment comprises sampling the audio environment to obtain sampled audio data in accordance with a first time interval, computing features of the sampled audio data, inferring an audio cluster identifier from the features of the sampled audio data in accordance with a second time interval, and updating an audio environment model using the features of the sampled audio data in accordance with a third time interval.

主权项

1. A method of classifying an audio environment, comprising:
sampling the audio environment with a microphone to obtain sampled audio data in accordance with a first time interval; computing features of the sampled audio data including computing a plurality of mel-frequency cepstral coefficients (MFCCs), generating running estimates of statistics of MFCC data over the first time interval using the plurality of MFCCs, storing the running estimates of statistics of MFCC data; inferring an audio cluster identifier from the features of the sampled audio data in accordance with a second time interval; and updating an audio environment model using the stored running estimates of statistics of MFCC data in accordance with a third time interval.