Multilingual bottle-neck feature learning from untranscribed speech

By Hongjie Chen, Cheung Chi Leung, Lei Xie, Bin Ma, Haizhou Li. 2017

31 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. James R. Glass
3. Wei-Ning Hsu
4. Aren Jansen
5. Ron J. Weiss
6. Herman Kamper
7. Haizhou Li
8. Aäron van den Oord

In Representation & Self-Supervised Learning

1. James R. Glass
2. Emmanuel Dupoux
3. Thomas Hain
4. Mingjie Chen
5. Aren Jansen
6. Herman Kamper
7. Wei-Ning Hsu
8. Maarten Versteegh

In Speech & Audio

1. Emmanuel Dupoux
2. James R. Glass
3. Aren Jansen
4. Maarten Versteegh
5. Haizhou Li
6. Wei-Ning Hsu
7. Thomas Hain
8. Thomas Schatz

In Signal Processing

1. James R. Glass
2. Emmanuel Dupoux
3. Maarten Versteegh
4. Aren Jansen
5. Haizhou Li
6. Aäron van den Oord
7. Ron J. Weiss
8. Samy Bengio

Citations in the corpus

• Unsupervised Speech Representation Learning Using WaveNet Autoencoders (2019, background): “This is also visible in the large degradation [57] AND THREE OTHER SUBMISSIONS THAT USED NEURAL NETWORK IN AN UNSUPERVISED SETTING [58], [59], [60].”
• The zero resource speech challenge 2017 (2017, methodology): “System [Y2] (not reported in [24]) obtained pairs from all corpora of all five languages.”
• Extracting bottleneck features and word-like pairs from untranscribed speech for feature representation (2017, methodology): “We follow the steps provided in [14], and use BNFs as our initial features for pairwise learning.”
• Measuring the perceptual availability of phonological features during language acquisition using unsupervised binary stochastic autoencoders (2019, background): “…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 al., 2017; Shibata et al., 2017; Ansari et al., 2017a,b) perform unsupervised ABX discrimination…”
• Unsupervised Acoustic Unit Representation Learning for Voice Conversion using WaveNet Auto-encoders (2020, methodology): “Besides, vocal tract length normalization (VTLN) [19] and fMLLR for speaker normalization [17] were also used.”
• Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling (2019, background): “[8] applied vocal tract length normalization (VTLN).”