Speaker identification made easy with pruned reassigned spectrograms
Sean A. Fulop, Youngwook Kim · The Journal of the Acoustical Society of America · 2013
One common scenario for speaker identification presents the task of identifying samples of speech from members of a previously enrolled group. One recent (and typical) set of results used 36 s of speech from each speaker to train Gaussian models by expectation-maximization during enrollment, and 20 s of speech for the test samples. Three major problems with this procedure are (1) sensitivity to noise; (2) impractical amounts of speech are required; (3) computationally expensive training is required. In our study, the reassigned spectrogram is pruned using phase-derivative indicator functions to provide a sparse time-frequency matrix of very small (40 ms) samples of speech. The pruning eliminates Gaussian noise up to ~1 dB SNR. Principle components analysis provided a set of 30 features from each spectrogram. Using an enrolled group of 24 speakers recorded under low-fidelity conditions, 83% identification accuracy (comparable to state of the art results with 6 dB SNR) was achieved from real-time classification methods (e.g., support vector machines) without need for extensive training. Moreover, these results extend to 1 dB SNR where standard techniques break down. The three main problems of speaker identification can thus be better addressed by our methodology.