Histogram Transform-based Speaker Identification
This work addresses speaker identification for security or voice recognition applications, but it appears incremental as it builds on existing feature extraction techniques with a new estimation method.
The paper tackles text-independent speaker identification by proposing a novel method that uses super-MFCCs features and histogram transform for probability density function estimation, showing promising improvement over conventional methods like Gaussian mixture models.
A novel text-independent speaker identification (SI) method is proposed. This method uses the Mel-frequency Cepstral coefficients (MFCCs) and the dynamic information among adjacent frames as feature sets to capture speaker's characteristics. In order to utilize dynamic information, we design super-MFCCs features by cascading three neighboring MFCCs frames together. The probability density function (PDF) of these super-MFCCs features is estimated by the recently proposed histogram transform~(HT) method, which generates more training data by random transforms to realize the histogram PDF estimation and recedes the commonly occurred discontinuity problem in multivariate histograms computing. Compared to the conventional PDF estimation methods, such as Gaussian mixture models, the HT model shows promising improvement in the SI performance.