Singing Voice Separation: A Study on Training Data
By Laure Prétet, Romain Hennequin, Jimena Royo-Letelier, Andrea Vaglio. 2019
33 citations recorded by the source corpus.
In the recent years, singing voice separation systems showed increased performance due to the use of supervised training. The design of training datasets is known as a crucial factor in the performance of such systems. We investigate on how the characteristics of the training dataset impacts the separation performances of state-of-the-art singing voice separation algorithms. We show that the separation quality and diversity are two important and complementary assets of a good training dataset. We also provide insights on possible transforms to perform data augmentation for this task.
Subfields: Artificial Intelligence, Signal Processing
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Alexandre D'efossez
- Yoshua Bengio
- L. Eon Bottou
- Jason Aaron Edward Weston
- Francis R. Bach
- Nicolas Usunier
- Martin Jaggi
- Mohit Bansal
In Signal Processing
- Alexandre D'efossez
- Francis R. Bach
- L. Eon Bottou
- Nicolas Usunier
- Félix Voituret
- Manuel Moussallam
- Anis Khlif
- Romain Hennequin
Citations in the corpus
- Music Source Separation in the Waveform Domain (2019, methodology): “Hennequin et al. [2020] propose another U-Net architecture for spectrogram masking, called Spleeter, that is trained on the Bean dataset Prétet et al. [2019], composed of short extracts from nearly 25,000 songs.”
- Spleeter: a fast and efficient music source separation tool with pre-trained models (2020, methodology): “The models were trained on Deezer’s internal datasets (noteworthily the Bean dataset that was used in (Prétet et al., 2019)) using Adam (Kingma & Ba, 2014).”
- SPLEETER: A FAST AND STATE-OF-THE ART MUSIC SOURCE SEPARATION TOOL WITH PRE-TRAINED MODELS (2019, methodology): “The pre-trained models are U-nets [2] and follows similar specifications as in [5].”