Sparse pursuit and dictionary learning for blind source separation in polyphonic music recordings

Autor: Sören Schulze, Emily J. King
Jazyk: angličtina
Rok vydání: 2021
Předmět:
Computer Science - Machine Learning
Acoustics and Ultrasonics
Computer science
lcsh:QC221-246
02 engineering and technology
Blind signal separation
Unsupervised learning
Computer Science - Sound
lcsh:QA75.5-76.95
030507 speech-language pathology & audiology
03 medical and health sciences
Pitch-invariance
0202 electrical engineering
electronic engineering
information engineering
Electrical and Electronic Engineering
Representation (mathematics)
Pattern matching
Audio signal
business.industry
Short-time Fourier transform
020206 networking & telecommunications
Pattern recognition
Dictionary learning
Feature (computer vision)
Computer Science::Sound
Inharmonicity
Parametric model
lcsh:Acoustics. Sound
Spectrogram
Blind source separation
Artificial intelligence
lcsh:Electronic computers. Computer science
0305 other medical science
business
Sparsity
Electrical Engineering and Systems Science - Audio and Speech Processing
Zdroj: EURASIP Journal on Audio, Speech, and Music Processing, Vol 2021, Iss 1, Pp 1-25 (2021)
ISSN: 1687-4722
Popis: We propose an algorithm for the blind separation of single-channel audio signals. It is based on a parametric model that describes the spectral properties of the sounds of musical instruments independently of pitch. We develop a novel sparse pursuit algorithm that can match the discrete frequency spectra from the recorded signal with the continuous spectra delivered by the model. We first use this algorithm to convert an STFT spectrogram from the recording into a novel form of log-frequency spectrogram whose resolution exceeds that of the mel spectrogram. We then make use of the pitch-invariant properties of that representation in order to identify the sounds of the instruments via the same sparse pursuit method. As the model parameters which characterize the musical instruments are not known beforehand, we train a dictionary that contains them, using a modified version of Adam. Applying the algorithm on various audio samples, we find that it is capable of producing high-quality separation results when the model assumptions are satisfied and the instruments are clearly distinguishable, but combinations of instruments with similar spectral characteristics pose a conceptual difficulty. While a key feature of the model is that it explicitly models inharmonicity, its presence can also still impede performance of the sparse pursuit algorithm. In general, due to its pitch-invariance, our method is especially suitable for dealing with spectra from acoustic instruments, requiring only a minimal number of hyperparameters to be preset. Additionally, we demonstrate that the dictionary that is constructed for one recording can be applied to a different recording with similar instruments without additional training.
Databáze: OpenAIRE
Externí odkaz: