Pescadores/Pescador: 1.1.0

McFee, Brian; Jacoby, Christopher; Humphrey, Eric J.; Pimenta, Waldir

doi:10.5281/zenodo.848831

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Models were implemented using Keras [70] and Tensor-Flow [71]. Each model was trained using the Adam optimizer [72], with data sampled using Pescador 1.1 [73]. Models were trained on mini-batches of 16 10-second patches.…”

Section: Training and Evaluationmentioning

confidence: 99%

Adaptive Pooling Operators for Weakly Labeled Sound Event Detection

McFee

Salamon

Bello

2018

IEEE/ACM Trans. Audio Speech Lang. Process.

144

130

View full text Add to dashboard Cite

Sound event detection (SED) methods are tasked with labeling segments of audio recordings by the presence of active sound sources. SED is typically posed as a supervised machine learning problem, requiring strong annotations for the presence or absence of each sound source at every time instant within the recording. However, strong annotations of this type are both labor-and cost-intensive for human annotators to produce, which limits the practical scalability of SED methods.In this work, we treat SED as a multiple instance learning (MIL) problem, where training labels are static over a short excerpt, indicating the presence or absence of sound sources but not their temporal locality. The models, however, must still produce temporally dynamic predictions, which must be aggregated (pooled) when comparing against static labels during training. To facilitate this aggregation, we develop a family of adaptive pooling operators-referred to as auto-pool-which smoothly interpolate between common pooling operators, such as min-, max-, or average-pooling, and automatically adapt to the characteristics of the sound sources in question. We evaluate the proposed pooling operators on three datasets, and demonstrate that in each case, the proposed methods outperform non-adaptive pooling operators for static prediction, and nearly match the performance of models trained with strong, dynamic annotations. The proposed method is evaluated in conjunction with convolutional neural networks, but can be readily applied to any differentiable model for time-series label prediction. While this article focuses on SED applications, the proposed methods are general, and could be applied widely to MIL problems in any domain.

show abstract

Section: Training and Evaluationmentioning

confidence: 99%