Librosa/Librosa: 0.6.1

McFee, Brian; McVicar, Matt; Balke, Stefan; Thomé, Carl; Raffel, Colin; Lee, Dana; Nieto, Oriol; Battenberg, Eric; Ellis, Dan; Yamamoto, Ryōichi; Moore, Josh; Bittner, Rachel M.; Choi, Keunwoo; Friesch, Pius; Stöter, Fabian-Robert; Lostanlen, Vincent; Waloschek, Simon; Seth,; Naktinis, Rimvydas; Repetto, Douglas; Hawthorne, Curtis Fjord; Carr, C. Jelleff; Pimenta, Waldir; Viktorin, Petr; Brossier, Paul; Santos, João Felipe; JackieWu,; Erik,; Holovaty, Adrian

doi:10.5281/zenodo.1252297

Cited by 6 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We denote this feature as WORD2VEC in the sequel. For ESC-50, by following [33], we use the librosa package [34] to compute mel-frequency cepstral coefficients (MFCC) for each clip.…”

Section: Dataset Preparationmentioning

confidence: 99%

Autoencoder based sample selection for self-taught learning

Feng

Duarte

2020

Knowledge-Based Systems

View full text Add to dashboard Cite

Self-taught learning is a technique that uses a large number of unlabeled data as source samples to improve the task performance on target samples. Compared with other transfer learning techniques, self-taught learning can be applied to a broader set of scenarios due to the loose restrictions on source data. However, knowledge transferred from source samples that are not sufficiently related to the target domain may negatively influence the target learner, which is referred to as negative transfer. In this paper, we propose a metric for the relevance between a source sample and target samples. To be more specific, both source and target samples are reconstructed through a single-layer autoencoder with a linear relationship between source samples and target samples simultaneously enforced. An 2,1 -norm sparsity constraint is imposed on the transformation matrix to identify source samples relevant to the target domain. Source domain samples that are deemed relevant are assigned pseudo-labels reflecting their relevance to target domain samples, and are combined with target samples in order to provide an expanded training set for classifier training. Local data structures are also preserved during source sample selection through spectral graph analysis. Promising results in extensive experiments show the advantages of the proposed approach.

show abstract

“…We denote this feature as WORD2VEC in the sequel. For ESC-50, by following [33], we use the librosa package [34] to compute mel-frequency cepstral coefficients (MFCC) for each clip.…”

Section: Dataset Preparationmentioning

confidence: 99%

Autoencoder based sample selection for self-taught learning

Feng

Duarte

2020

Knowledge-Based Systems

View full text Add to dashboard Cite

show abstract

“…First, the audio file needs to be preprocessed. For this purpose, we use the python package librosa (McFee et al, 2018). First, we convert the audio file to mono.…”

Section: Preprocessingmentioning

confidence: 99%

DECIBEL: Improving Audio Chord Estimation for Popular Music by Alignment and Integration of Crowd-Sourced Symbolic Representations

Odekerken,

Koops,

Volk

2020

Preprint

View full text Add to dashboard Cite

Automatic Chord Estimation (ace) is a fundamental task in Music Information Retrieval (mir), which has applications in both music performance and in research on other mir tasks. The ace task consists of segmenting a music recording or score and assigning a chord label to each segment. Ace has been a task in the mirex competition for 10 years, but is not yet a solved problem: current methods seem to have reached a glass ceiling. Moreover, many recent methods are trained on a limited data set and consequently suffer from overfitting. In this study, we propose decibel 1 (DEtection of Chords Improved By Exploiting Linking symbolic formats), a novel system that utilizes multiple symbolic music representations in addition to audio in order to improve ace on popular music.

show abstract

“…The choice of minimal frequency at 2 kHz corresponds to a lower bound on the vocal range of avian flight calls of thrushes. With the librosa Python library [88], the computation of logmelspec is about 20 times faster than real time on a dual-core Intel Xeon E-2690v2 3.0 GHz central processing unit (CPU).…”

Section: Baseline: Convolutional Neural Networkmentioning

confidence: 99%

Robust sound event detection in bioacoustic sensor networks

et al. 2019

Self Cite

View full text Add to dashboard Cite

Bioacoustic sensors, sometimes known as autonomous recording units (ARUs), can record sounds of wildlife over long periods of time in scalable and minimally invasive ways. Deriving per-species abundance estimates from these sensors requires detection, classification, and quantification of animal vocalizations as individual acoustic events. Yet, variability in ambient noise, both over time and across sensors, hinders the reliability of current automated systems for sound event detection (SED), such as convolutional neural networks (CNN) in the time-frequency domain. In this article, we develop, benchmark, and combine several machine listening techniques to improve the generalizability of SED models across heterogeneous acoustic environments. As a case study, we consider the problem of detecting avian flight calls from a ten-hour recording of nocturnal bird migration, recorded by a network of six ARUs in the presence of heterogeneous background noise. Starting from a CNN yielding state-of-the-art accuracy on this task, we introduce two noise adaptation techniques, respectively integrating short-term (60 ms) and long-term (30 min) context. First, we apply per-channel energy normalization (PCEN) in the time-frequency domain, which applies short-term automatic gain control to every subband in the mel-frequency spectrogram. Secondly, we replace the last dense layer in the network by a context-adaptive neural network (CA-NN) layer, i.e. an affine layer whose weights are dynamically adapted at prediction time by an auxiliary network taking long-term summary statistics of spectrotemporal features as input. We show that PCEN reduces temporal overfitting across dawn vs. dusk audio clips whereas context adaptation on PCEN-based summary statistics reduces spatial overfitting across sensor locations. Moreover, combining them yields state-of-the-art results that are unmatched by artificial data augmentation alone. We release a pre-trained version of our best performing system under the name of BirdVoxDetect, a ready-to-use detector of avian flight calls in field recordings.

show abstract

Librosa/Librosa: 0.6.1

Cited by 6 publications

References 0 publications

Autoencoder based sample selection for self-taught learning

Autoencoder based sample selection for self-taught learning

DECIBEL: Improving Audio Chord Estimation for Popular Music by Alignment and Integration of Crowd-Sourced Symbolic Representations

Robust sound event detection in bioacoustic sensor networks

Contact Info

Product

Resources

About