Unsupervised Classification of Voiced Speech and Pitch Tracking Using Forward-Backward Kalman Filtering
This addresses speech processing challenges for applications like noise-robust analysis, though it is incremental as it combines existing techniques.
The paper tackles the problem of detecting voiced speech, estimating fundamental frequency, and tracking pitch in noisy environments by integrating these subtasks into a single algorithm, showing favorable comparison with state-of-the-art pitch detection methods.
The detection of voiced speech, the estimation of the fundamental frequency, and the tracking of pitch values over time are crucial subtasks for a variety of speech processing techniques. Many different algorithms have been developed for each of the three subtasks. We present a new algorithm that integrates the three subtasks into a single procedure. The algorithm can be applied to pre-recorded speech utterances in the presence of considerable amounts of background noise. We combine a collection of standard metrics, such as the zero-crossing rate, for example, to formulate an unsupervised voicing classifier. The estimation of pitch values is accomplished with a hybrid autocorrelation-based technique. We propose a forward-backward Kalman filter to smooth the estimated pitch contour. In experiments, we are able to show that the proposed method compares favorably with current, state-of-the-art pitch detection algorithms.