A deep scattering spectrum — Deep Siamese network pipeline for unsupervised acoustic modeling

By Neil Zeghidour, Gabriel Synnaeve, Maarten Versteegh, Emmanuel Dupoux. 2016

39 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Representation & Self-Supervised Learning, Speech & Audio, 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

1. Emmanuel Dupoux
2. Eugene Belilovsky
3. Thomas Schatz
4. Edouard Oyallon
5. Michael Eickenberg
6. Neil Zeghidour
7. Ewan Dunbar
8. Maarten Versteegh

In Representation & Self-Supervised Learning

1. Joan Bruna
2. Edouard Oyallon
3. Eugene Belilovsky
4. Matthew J. Hirn
5. Tomás Angles
6. Stéphane Mallat
7. Joakim Andén
8. Neil Zeghidour

In Speech & Audio

1. Emmanuel Dupoux
2. Thomas Schatz
3. Neil Zeghidour
4. Ewan Dunbar
5. Maarten Versteegh
6. Rahma Chaabouni
7. Eugene Belilovsky
8. Edouard Oyallon

In Signal Processing

1. Emmanuel Dupoux
2. Thomas Schatz
3. Neil Zeghidour
4. Maarten Versteegh
5. Stéphane Mallat
6. Joan Bruna
7. Edouard Oyallon
8. Ewan Dunbar

Citations in the corpus

• Learning Weakly Supervised Multimodal Phoneme Embeddings (2017, methodology): “This was already demonstrated for the acoustic modality [6, 27, 28]; we demonstrate that ABnet also improves the visual and concatenated input representations.”
• Kymatio: Scattering Transforms in Python (2018, background): “It is also effective in combination with modern representation learning approaches (Oyallon et al., 2018; Sainath et al., 2014; Zeghidour et al., 2016).”
• ABX-Discriminability Measures and Applications. (Mesures de discriminabilité ABX et applications) (2016, background): “A study published after the initial session of the challenge [103] also showed that replacing the original input features (Mel-filterbanks) with richer scattering spectrum coefficients [104] improved the performance of Siamese Network acoustic models (17.”
• Measuring the perceptual availability of phonological features during language acquisition using unsupervised binary stochastic autoencoders (2019, background): “…al., 2015; Renshaw et al., 2015; Agenbag and Niesler, 2015; Chen et al., 2015; Baljekar et al., 2015; Räsänen et al., 2015; Lyzinski et al., 2015; Zeghidour et al., 2016; Heck et al., 2016; Srivastava and Shrivastava, 2016; Kamper et al., 2017b; Chen et al., 2017; Yuan et al., 2017; Heck et…”
• Learning Text Similarity with Siamese Recurrent Networks (2016, background): “In other areas, Siamese networks have been applied to such diverse tasks as unsupervised acoustic modelling (Synnaeve et al., 2014; Thiolliere et al., 2015; Kamper et al., 2016; Zeghidour et al., 2015), learning food preferences (Yang et al., 2015) and scene detection (Baraldi et al., 2015).”