61 citations
Counting papers with at most 25 authors, so collaboration author lists do not stand in for the people here.
- Université Paris CitéFR6 papers
- Laboratoire de Sciences Cognitives et PsycholinguistiqueFR5 papers
- Centre National de la Recherche ScientifiqueFR4 papers
- Institut National des Langues et Civilisations OrientalesFR3 papers
- Laboratoire d'Informatique de GrenobleFR3 papers
- Langues et Civilisations à Tradition OraleFR3 papers
- Sorbonne Paris CitéFR3 papers
- Brno University of TechnologyCZ2 papers
- Meta (Israel)IL2 papers
- Meta (United States)US2 papers
- Nara Institute of Science and TechnologyJP2 papers
- University of TorontoCA2 papers
19 papers
The Unheard Alternative: Contrastive Explanations for Speech-to-Text Models
Lina Conti, Dennis Fucci, Marco Gaido +3
Contrastive explanations, which indicate why an AI system produced one output (the target) instead of another (the foil), are widely regarded in explainable AI as more informative…
A solution to the mystery of the sub-harmonic series via a linear model of the cochlea
Ugo Boscain, Xiangyu Ma, Dario Prandi +1
In this paper, we study a simple linear model of the cochlea as a set of vibrating strings. We make hypothesis that the information sent to the auditory cortex is the energy stored…
Assessing the Impact of Anisotropy in Neural Representations of Speech: A Case Study on Keyword Spotting
Guillaume Wisniewski, Séverine Guillaume, Clara Rosina Fernández
Pretrained speech representations like wav2vec2 and HuBERT exhibit strong anisotropy, leading to high similarity between random embeddings. While widely observed, the impact of thi…
Probing structural constraints of negation in Pretrained Language Models
David Kletz, Marie Candito, Pascal Amsili
Contradictory results about the encoding of the semantic impact of negation in pretrained language models (PLMs). have been drawn recently (e.g. Kassner and Sch{ü}tze (2020); Gubel…
Establishing degrees of closeness between audio recordings along different dimensions using large-scale cross-lingual models
Maxime Fily, Guillaume Wisniewski, Severine Guillaume +2
In the highly constrained context of low-resource language studies, we explore vector representations of speech from a pretrained model to determine their level of abstraction with…
Character Recognition in Byzantine Seals with Deep Neural Networks
Théophile Rageau, Laurence Likforman-Sulem, Attilio Fiandrotti +3
Seals are small coin-shaped artifacts, mostly made of lead, held with strings to seal letters. This work presents the first attempt towards automatic reading of text on Byzantine s…
From `Snippet-lects' to Doculects and Dialects: Leveraging Neural Representations of Speech for Placing Audio Signals in a Language Landscape
Séverine Guillaume, Guillaume Wisniewski, Alexis Michaud
XLSR-53 a multilingual model of speech, builds a vector representation from audio, which allows for a range of computational treatments. The experiments reported here use this neur…
Assessing the Capacity of Transformer to Abstract Syntactic Representations: A Contrastive Analysis Based on Long-distance Agreement
Bingzhi Li, Guillaume Wisniewski, Benoît Crabbé
The long-distance agreement, evidence for syntactic structure, is increasingly used to assess the syntactic generalization of Neural Language Models. Much work has shown that trans…
Are word boundaries useful for unsupervised language learning?
Tu Anh Nguyen, Maureen de Seyssel, Robin Algayres +3
Word or word-fragment based Language Models (LM) are typically preferred over character-based ones in many downstream applications. This may not be surprising as words seem more li…
Toward a realistic model of speech processing in the brain with self-supervised learning
Juliette Millet, Charlotte Caucheteux, Pierre Orhan +5
Several deep neural networks have recently been shown to generate activations similar to those of the brain in response to the same input. These algorithms, however, remain largely…
Probing phoneme, language and speaker information in unsupervised speech representations
Maureen de Seyssel, Marvin Lavechin, Yossi Adi +2
Unsupervised models of representations based on Contrastive Predictive Coding (CPC)[1] are primarily used in spoken language modelling in that they encode phonetic information. In…
Not quite there yet: Combining analogical patterns and encoder-decoder networks for cognitively plausible inflection
Basilio Calderone, Nabil Hathout, Olivier Bonami
The paper presents four models submitted to Part 2 of the SIGMORPHON 2021 Shared Task 0, which aims at replicating human judgements on the inflection of nonce lexemes. Our goal is…
User-friendly automatic transcription of low-resource languages: Plugging ESPnet into Elpis
Oliver Adams, Benjamin Galliot, Guillaume Wisniewski +11
This paper reports on progress integrating the speech recognition toolkit ESPnet into Elpis, a web front-end originally designed to provide access to the Kaldi automatic speech rec…
The Zero Resource Speech Challenge 2020: Discovering discrete subword and word units
Ewan Dunbar, Julien Karadayi, Mathieu Bernard +6
We present the Zero Resource Speech Challenge 2020, which aims at learning speech representations from raw audio signals without any labels. It combines the data sets and metrics f…
The Perceptimatic English Benchmark for Speech Perception Models
Juliette Millet, Ewan Dunbar
We present the Perceptimatic English Benchmark, an open experimental benchmark for evaluating quantitative models of speech perception in English. The benchmark consists of ABX sti…
FlauBERT: Unsupervised Language Model Pre-training for French
Hang Le, Loïc Vial, Jibril Frej +7
Language models have become a key step to achieve state-of-the art results in many different Natural Language Processing (NLP) tasks. Leveraging the huge amount of unlabeled texts…
The Zero Resource Speech Challenge 2019: TTS without T
Ewan Dunbar, Robin Algayres, Julien Karadayi +10
We present the Zero Resource Speech Challenge 2019, which proposes to build a speech synthesizer without any text or phonetic labels: hence, TTS without T (text-to-speech without t…
RNNs Implicitly Implement Tensor Product Representations
R. Thomas McCoy, Tal Linzen, Ewan Dunbar +1
Recurrent neural networks (RNNs) can learn continuous vector representations of symbolic structures such as sequences and sentences; these representations often exhibit linear regu…
Learning to detect dysarthria from raw speech
Juliette Millet, Neil Zeghidour
Speech classifiers of paralinguistic traits traditionally learn from diverse hand-crafted low-level features, by selecting the relevant information for the task at hand. We explore…