A multi-pitch tracking algorithm for noisy speech

By Mingyang Wu, Deliang Wang, Guy J. Brown. IEEE International Conference on Acoustics, Speech, and Signal Processing, 2002

328 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Speech segregation based on sound localization (2001, methodology): “For the two-source scenario, one test set is the corpus collected by Cooke (1993), chosen because it is commonly used in computational ASA studies (Brown and Cooke, 1994; Wang and Brown, 1999; Wu et al., 2002).”
Blind One-microphone Speech Separation: A Spectral Learning Approach (2004, methodology): “If S pitches are sought, the output that we need to obtain from the pitch extractor is, for each time frame n, the S pitches !n1,...,!nS, as well as the strength ynms of the s-th pitch for each frequency m. Several pitch extractors for multiple pitches may be used [ 14 , 15].”
An Overview of Speaker Identification: Accuracy and Robustness Issues (2011, result): “Weiss and Ellis also compared the performance of the produced RVM masks to CASA pitch [105] based masking for speech recognition using reconstruction.”
Deep Belief Networks Based Voice Activity Detection (2013, methodology): “Note that although the pitch detection technique has been greatly improved in recent years [39], [49]–[54], we extract pitch features from the noisy speech signals by an early subharmonic-to-harmonic-ratio-based pitch detection algorithm [55]”
A pitch extraction algorithm tuned for automatic speech recognition (2014, methodology): “We started our work by obtaining various off-the-shelf pitch extractors, namely Yin [2], Getf0 [3], SAcC [4], Wu [5], SWIPE [6] and YAAPT [7].”