1 |
LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech
|
|
|
|
In: INTERSPEECH 2021: Conference of the International Speech Communication Association ; https://hal.archives-ouvertes.fr/hal-03317730 ; INTERSPEECH 2021: Conference of the International Speech Communication Association, Aug 2021, Brno, Czech Republic (2021)
|
|
BASE
|
|
Show details
|
|
2 |
LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech
|
|
|
|
In: INTERSPEECH 2021: ; INTERSPEECH 2021: Conference of the International Speech Communication Association ; https://hal.archives-ouvertes.fr/hal-03317730 ; INTERSPEECH 2021: Conference of the International Speech Communication Association, Aug 2021, Brno, Czech Republic (2021)
|
|
BASE
|
|
Show details
|
|
3 |
LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech
|
|
Evain, Solène; Nguyen, Ha; Le, Hang; Zanon Boito, Marcely; Mdhaffar, Salima; Alisamir, Sina; Tong, Ziyi; Tomashenko, Natalia; Dinarelli, Marco; Parcollet, Titouan; Allauzen, Alexandre; Estève, Yannick; Lecouteux, Benjamin; Portet, François; Rossato, Solange; Ringeval, Fabien; Schwab, Didier; Besacier, Laurent
|
|
In: INTERSPEECH 2021: ; INTERSPEECH 2021: Conference of the International Speech Communication Association ; https://hal.archives-ouvertes.fr/hal-03317730 ; INTERSPEECH 2021: Conference of the International Speech Communication Association, Aug 2021, Brno, Czech Republic (2021)
|
|
Abstract:
International audience ; Self-Supervised Learning (SSL) using huge unlabeled data has been successfully explored for image and natural language processing. Recent works also investigated SSL from speech. They were notably successful to improve performance on downstream tasks such as automatic speech recognition (ASR). While these works suggest it is possible to reduce dependence on labeled data for building efficient speech systems, their evaluation was mostly made on ASR and using multiple and heterogeneous experimental settings (most of them for English). This questions the objective comparison of SSL approaches and the evaluation of their impact on building speech systems. In this paper, we propose LeBenchmark: a reproducible framework for assessing SSL from speech. It not only includes ASR (high and low resource) tasks but also spoken language understanding, speech translation and emotion recognition. We also focus on speech technologies in a language different than English: French. SSL models of different sizes are trained from carefully sourced and documented datasets. Experiments show that SSL is beneficial for most but not all tasks which confirms the need for exhaustive and reliable benchmarks to evaluate its real impact. LeBenchmark is shared with the scientific community for reproducible research in SSL from speech.
|
|
Keyword:
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]; ASR; Automatic Emotion Recognition; Self-Supervised Representation Learning; SLU; Speech Translation
|
|
URL: https://hal.archives-ouvertes.fr/hal-03317730v2/file/FLOWBERT_IS2021.pdf https://hal.archives-ouvertes.fr/hal-03317730v2/document https://hal.archives-ouvertes.fr/hal-03317730
|
|
BASE
|
|
Hide details
|
|
4 |
The contribution of visual articulatory gestures and orthography to speech processing: Evidence from novel word learning
|
|
|
|
In: ISSN: 0278-7393 ; EISSN: 1939-1285 ; Journal of Experimental Psychology: Learning, Memory, and Cognition ; https://hal.archives-ouvertes.fr/hal-03189083 ; Journal of Experimental Psychology: Learning, Memory, and Cognition, American Psychological Association, In press, ⟨10.1037/xlm0001036⟩ (2021)
|
|
BASE
|
|
Show details
|
|
5 |
Learning robust speech representation with an articulatory-regularized variational autoencoder
|
|
|
|
In: Proccedings of Interspeech 2021 ; Interspeech 2021 - 22nd Annual Conference of the International Speech Communication Association ; https://hal.archives-ouvertes.fr/hal-03373252 ; Interspeech 2021 - 22nd Annual Conference of the International Speech Communication Association, Aug 2021, Brno, Czech Republic (2021)
|
|
BASE
|
|
Show details
|
|
6 |
Do Infants Really Learn Phonetic Categories?
|
|
|
|
In: EISSN: 2470-2986 ; Open Mind ; https://hal.archives-ouvertes.fr/hal-03550830 ; Open Mind, MIT Press, 2021, 5, pp.113-131. ⟨10.1162/opmi_a_00046⟩ (2021)
|
|
BASE
|
|
Show details
|
|
9 |
A multispeaker dataset of raw and reconstructed speech production real-time MRI video and 3D volumetric images ...
|
|
|
|
BASE
|
|
Show details
|
|
11 |
A multispeaker dataset of raw and reconstructed speech production real-time MRI video and 3D volumetric images ...
|
|
|
|
BASE
|
|
Show details
|
|
13 |
Models and activations for "Can phones, syllables, and words emerge as side-products of cross-situational audiovisual learning? - A computational investigation" ...
|
|
|
|
BASE
|
|
Show details
|
|
14 |
Models and activations for "Can phones, syllables, and words emerge as side-products of cross-situational audiovisual learning? - A computational investigation" ...
|
|
|
|
BASE
|
|
Show details
|
|
15 |
Manual praxis and language-production networks: An fMRI dataset ...
|
|
|
|
BASE
|
|
Show details
|
|
16 |
Can phones, syllables, and words emerge as side-products of cross-situational audiovisual learning? - A computational investigation ...
|
|
|
|
BASE
|
|
Show details
|
|
17 |
Le discours médiatique comme relation de pouvoir symbolique : pratiques de médiatisation de la diaspora
|
|
|
|
In: Argumentum: Journal of the Seminar of Discursive Logic, Argumentation Theory and Rhetoric, Vol 19, Iss 1, Pp 45-65 (2021) (2021)
|
|
BASE
|
|
Show details
|
|
18 |
Towards unsupervised learning of speech features in the wild
|
|
|
|
In: SLT 2020 : IEEE Spoken Language Technology Workshop ; https://hal.archives-ouvertes.fr/hal-03070411 ; SLT 2020 : IEEE Spoken Language Technology Workshop, Dec 2020, Shenzhen / Virtual, China (2020)
|
|
BASE
|
|
Show details
|
|
19 |
Evaluating the reliability of acoustic speech embeddings
|
|
|
|
In: INTERSPEECH 2020 - Annual Conference of the International Speech Communication Association ; https://hal.inria.fr/hal-02977539 ; INTERSPEECH 2020 - Annual Conference of the International Speech Communication Association, Oct 2020, Shanghai / Vitrtual, China (2020)
|
|
BASE
|
|
Show details
|
|
20 |
Data Augmenting Contrastive Learning of Speech Representations in the Time Domain
|
|
|
|
In: SLT 2020 - IEEE Spoken Language Technology Workshop ; https://hal.archives-ouvertes.fr/hal-03070321 ; SLT 2020 - IEEE Spoken Language Technology Workshop, Dec 2020, Shenzhen / Virtual, China (2020)
|
|
BASE
|
|
Show details
|
|
|
|