An overview of lead and accompaniment separation in music
Popular music is often composed of an accompaniment and a lead component, the latter
typically consisting of vocals. Filtering such mixtures to extract one or both components has …
typically consisting of vocals. Filtering such mixtures to extract one or both components has …
[图书][B] Audio source separation and speech enhancement
Learn the technology behind hearing aids, Siri, and Echo Audio source separation and
speech enhancement aim to extract one or more source signals of interest from an audio …
speech enhancement aim to extract one or more source signals of interest from an audio …
Multichannel extensions of non-negative matrix factorization with complex-valued data
This paper presents new formulations and algorithms for multichannel extensions of non-
negative matrix factorization (NMF). The formulations employ Hermitian positive semidefinite …
negative matrix factorization (NMF). The formulations employ Hermitian positive semidefinite …
Hawkes processes for events in social media
This chapter provides an accessible introduction for point processes, and especially Hawkes
processes, for modeling discrete, inter-dependent events over continuous time. We start by …
processes, for modeling discrete, inter-dependent events over continuous time. We start by …
Deep learning for video classification and captioning
Deep learning for video classification and captioning Page 1 IPART MULTIMEDIA
CONTENT ANALYSIS Page 2 Page 3 1Deep Learning for Video Classification and …
CONTENT ANALYSIS Page 2 Page 3 1Deep Learning for Video Classification and …
Static and dynamic source separation using nonnegative factorizations: A unified view
Source separation models that make use of nonnegativity in their parameters have been
gaining increasing popularity in the last few years, spawning a significant number of …
gaining increasing popularity in the last few years, spawning a significant number of …
An unsupervised approach to cochannel speech separation
Cochannel (two-talker) speech separation is predominantly addressed using pretrained
speaker dependent models. In this paper, we propose an unsupervised approach to …
speaker dependent models. In this paper, we propose an unsupervised approach to …
A non-negative approach to semi-supervised separation of speech from noise with the use of temporal dynamics
GJ Mysore, P Smaragdis - 2011 IEEE International Conference …, 2011 - ieeexplore.ieee.org
We present a semi-supervised source separation methodology to denoise speech by
modeling speech as one source and noise as the other source. We model speech using the …
modeling speech as one source and noise as the other source. We model speech using the …
Deep polyphonic ADSR piano note transcription
We investigate a late-fusion approach to piano transcription, combined with a strong
temporal prior in the form of a handcrafted Hidden Markov Model (HMM). The network …
temporal prior in the form of a handcrafted Hidden Markov Model (HMM). The network …
Compositional models for audio processing: Uncovering the structure of sound mixtures
Many classes of data are composed as constructive combinations of parts. By constructive
combination, we mean additive combination that does not result in subtraction or …
combination, we mean additive combination that does not result in subtraction or …