Alert button
Picture for Ewan Dunbar

Ewan Dunbar

Alert button

Bigger is not Always Better: The Effect of Context Size on Speech Pre-Training

Dec 03, 2023
Sean Robertson, Ewan Dunbar

Figure 1 for Bigger is not Always Better: The Effect of Context Size on Speech Pre-Training
Figure 2 for Bigger is not Always Better: The Effect of Context Size on Speech Pre-Training
Figure 3 for Bigger is not Always Better: The Effect of Context Size on Speech Pre-Training
Figure 4 for Bigger is not Always Better: The Effect of Context Size on Speech Pre-Training
Viaarxiv icon

Zero Resource Code-switched Speech Benchmark Using Speech Utterance Pairs For Multiple Spoken Languages

Oct 04, 2023
Kuan-Po Huang, Chih-Kai Yang, Yu-Kuan Fu, Ewan Dunbar, Hung-yi Lee

Figure 1 for Zero Resource Code-switched Speech Benchmark Using Speech Utterance Pairs For Multiple Spoken Languages
Figure 2 for Zero Resource Code-switched Speech Benchmark Using Speech Utterance Pairs For Multiple Spoken Languages
Figure 3 for Zero Resource Code-switched Speech Benchmark Using Speech Utterance Pairs For Multiple Spoken Languages
Figure 4 for Zero Resource Code-switched Speech Benchmark Using Speech Utterance Pairs For Multiple Spoken Languages
Viaarxiv icon

Evaluating context-invariance in unsupervised speech representations

Oct 27, 2022
Mark Hallap, Emmanuel Dupoux, Ewan Dunbar

Figure 1 for Evaluating context-invariance in unsupervised speech representations
Figure 2 for Evaluating context-invariance in unsupervised speech representations
Figure 3 for Evaluating context-invariance in unsupervised speech representations
Figure 4 for Evaluating context-invariance in unsupervised speech representations
Viaarxiv icon

Self-supervised language learning from raw audio: Lessons from the Zero Resource Speech Challenge

Oct 27, 2022
Ewan Dunbar, Nicolas Hamilakis, Emmanuel Dupoux

Figure 1 for Self-supervised language learning from raw audio: Lessons from the Zero Resource Speech Challenge
Figure 2 for Self-supervised language learning from raw audio: Lessons from the Zero Resource Speech Challenge
Figure 3 for Self-supervised language learning from raw audio: Lessons from the Zero Resource Speech Challenge
Figure 4 for Self-supervised language learning from raw audio: Lessons from the Zero Resource Speech Challenge
Viaarxiv icon

Are word boundaries useful for unsupervised language learning?

Oct 06, 2022
Tu Anh Nguyen, Maureen de Seyssel, Robin Algayres, Patricia Roze, Ewan Dunbar, Emmanuel Dupoux

Figure 1 for Are word boundaries useful for unsupervised language learning?
Figure 2 for Are word boundaries useful for unsupervised language learning?
Figure 3 for Are word boundaries useful for unsupervised language learning?
Figure 4 for Are word boundaries useful for unsupervised language learning?
Viaarxiv icon

Toward a realistic model of speech processing in the brain with self-supervised learning

Jun 03, 2022
Juliette Millet, Charlotte Caucheteux, Pierre Orhan, Yves Boubenec, Alexandre Gramfort, Ewan Dunbar, Christophe Pallier, Jean-Remi King

Figure 1 for Toward a realistic model of speech processing in the brain with self-supervised learning
Figure 2 for Toward a realistic model of speech processing in the brain with self-supervised learning
Figure 3 for Toward a realistic model of speech processing in the brain with self-supervised learning
Figure 4 for Toward a realistic model of speech processing in the brain with self-supervised learning
Viaarxiv icon

Predicting non-native speech perception using the Perceptual Assimilation Model and state-of-the-art acoustic models

May 31, 2022
Juliette Millet, Ioana Chitoran, Ewan Dunbar

Figure 1 for Predicting non-native speech perception using the Perceptual Assimilation Model and state-of-the-art acoustic models
Figure 2 for Predicting non-native speech perception using the Perceptual Assimilation Model and state-of-the-art acoustic models
Figure 3 for Predicting non-native speech perception using the Perceptual Assimilation Model and state-of-the-art acoustic models
Figure 4 for Predicting non-native speech perception using the Perceptual Assimilation Model and state-of-the-art acoustic models
Viaarxiv icon

Do self-supervised speech models develop human-like perception biases?

May 31, 2022
Juliette Millet, Ewan Dunbar

Figure 1 for Do self-supervised speech models develop human-like perception biases?
Figure 2 for Do self-supervised speech models develop human-like perception biases?
Figure 3 for Do self-supervised speech models develop human-like perception biases?
Figure 4 for Do self-supervised speech models develop human-like perception biases?
Viaarxiv icon

The Interspeech Zero Resource Speech Challenge 2021: Spoken language modelling

Apr 29, 2021
Ewan Dunbar, Mathieu Bernard, Nicolas Hamilakis, Tu Anh Nguyen, Maureen de Seyssel, Patricia Rozé, Morgane Rivière, Eugene Kharitonov, Emmanuel Dupoux

Figure 1 for The Interspeech Zero Resource Speech Challenge 2021: Spoken language modelling
Figure 2 for The Interspeech Zero Resource Speech Challenge 2021: Spoken language modelling
Viaarxiv icon

Paraphrases do not explain word analogies

Feb 23, 2021
Louis Fournier, Ewan Dunbar

Figure 1 for Paraphrases do not explain word analogies
Figure 2 for Paraphrases do not explain word analogies
Figure 3 for Paraphrases do not explain word analogies
Figure 4 for Paraphrases do not explain word analogies
Viaarxiv icon