Sorry, you need to enable JavaScript to visit this website.

DEEPF0: END-TO-END FUNDAMENTAL FREQUENCY ESTIMATION FOR MUSIC AND SPEECH SIGNALS

Primary tabs

Citation Author(s):
Satwinder Singh, Ruili Wang, Yuanhang Qiu
Submitted by:
Satwinder singh
Last updated:
17 June 2021 - 7:48pm
Document Type:
Poster
Document Year:
2021
Event:
Presenters Name:
Satwinder Singh
Paper Code:
AUD-3.1

Abstract 

Abstract: 

We propose a novel pitch estimation technique called DeepF0, which leverages the available annotated data to directly learns from the raw audio in a data-driven manner. F0 estimation is important in various speech processing and music information retrieval applications. Existing deep learning models for pitch estimations have relatively limited learning capabilities due to their shallow receptive field. The proposed model addresses this issue by extending the receptive field of a network by introducing the dilated convolutional blocks into the network. The dilation factor increases the network receptive field exponentially without increasing the parameters of the model exponentially. To make the training process more efficient and faster, DeepF0 is augmented with residual blocks with residual connections. Our empirical evaluation demonstrates that the proposed model outperforms the baselines in terms of raw pitch accuracy and raw chroma accuracy even using 77.4% fewer network parameters. We also show that our model can capture reasonably well pitch estimation even under the various levels of accompaniment noise.

up
0 users have voted:

Dataset Files

ICASSP 2021 Poster

(58)

ICASSP 2021 Presentation

(64)