<div dir="ltr"><div><div class="gmail-grand-item-offre">
<div class="gmail-item-offre">
<p class="gmail-list-group-item-text"><b>Contract type : </b>
Fixed-term contract
</p>
</div>
<div class="gmail-item-offre">
<p class="gmail-list-group-item-text"><b>Level of qualifications required : </b>
Graduate degree or equivalent
</p>
</div>
<div class="gmail-item-offre">
<p class="gmail-list-group-item-text"><b>Fonction : </b>
PhD Position
</p>
</div>
</div>
<div class="gmail-grand-item-offre">
<h2 class="gmail-h2-detail gmail-list-group-item-heading">Context</h2>
<div class="gmail-list-group-item-text gmail-offre-detail-affiche-context">
<p><strong> Introduction</strong><br>Stuttering, a fluency disorder
affecting millions of individuals, is characterized by stuttering-like
disfluencies (blocks, prolongations, repetitions) linked to dysfunctions
in speech motor control. While its automatic detection has already been
explored using audio-based models, current systems remain limited by
low robustness, difficulty in identifying certain disfluencies such as
silent blocks, and reliance on scarce data. This PhD project proposes a
multimodal approach (audio, video, text) to enhance the accuracy and
robustness of disfluency detection, leveraging an audiovisual corpus of
French-speaking individuals who stutter. The analysis will rely on
modality-specific encoding techniques, followed by a strategic fusion of
their representations for final classification.</p>
<p><strong>Aims</strong></p>
<p>The aim of this PhD is to design, develop, and evaluate a multimodal
deep learning approach for the automatic detection of stuttering-like
disfluencies in French, by combining audio, video, and textual
modalities. The work will be based on an annotated audiovisual corpus of
French-speaking people who stutter, with particular focus on
disfluencies that are difficult to detect through audio alone, such as
silent blocks, and on robustness to individual variability.</p>
</div>
</div>
<div class="gmail-grand-item-offre">
<h2 class="gmail-h2-detail gmail-list-group-item-heading">Assignment</h2>
<div class="gmail-list-group-item-text gmail-offre-detail-affiche-context">
<p><strong>Missions</strong></p>
<p style="font-weight:400">The doctoral candidate’s work will include the following tasks:</p>
<ul style="font-weight:400"><li><strong>Audio encoding</strong>: Implement and adapt Stutternet (Sheikh, S. A., Sahidullah, M., Hirsch, F., & Ouni, S. – 2021 – <em>Stutternet: Stuttering detection using time delay neural network</em>, in EUSIPCO) to extract acoustic features relevant to disfluency detection by capturing temporal dependencies.</li><li><strong>Video encoding</strong>: Develop and train vision models
(e.g., C3D or Transformers) to analyze video sequences for visual cues
of stuttering (facial tension, blinking, atypical movements). The
extraction of facial landmarks (with OpenFace or MediaPipe) will also be
explored as a complementary or alternative source of features.</li><li><strong>Text encoding</strong>: Generate automatic transcriptions
(via Whisper) and encode them using pre-trained language models (BERT,
RoBERTa) to extract linguistic context and identify textual patterns
characteristic of disfluencies.</li><li><strong>Multimodal fusion</strong>: Implement and compare several
strategies to fuse the representations from the three modalities, such
as concatenation, adaptive attention mechanisms, or other approaches
leveraging data complementarity.</li><li><strong>Classification and evaluation</strong>: Develop a classifier
operating on the fused representation to predict the presence or
absence of stuttering within a given time window. Evaluation will rely
on standard metrics (precision, recall, F1-score, AUC), and results will
be compared to expert manual annotations. Qualitative analyses will
also be conducted to interpret model errors and refine the approach.</li></ul>
<p style="font-weight:400">Beyond detection, this PhD aims to
contribute methodologically to the field of multimodal fusion applied to
pathological speech, with potential impact in clinical contexts.</p>
</div>
</div>
<div class="gmail-grand-item-offre">
<h2 class="gmail-h2-detail gmail-list-group-item-heading">Main activities</h2>
<div class="gmail-list-group-item-text gmail-offre-detail-affiche-context">
<p><strong><span lang="EN-US">Required Skills</span></strong><span lang="EN-US"><br>The
candidate should hold a Master’s degree in computer science, have
strong skills in machine learning and deep learning, and be proficient
in Python and frameworks such as PyTorch or TensorFlow. An interest in
signal processing (audio/video) and ideally in NLP is expected.
Autonomy, rigor, critical thinking, and analytical abilities are
essential, along with strong communication skills to work in a
multidisciplinary environment. An interest in phonetics, linguistics,
and speech disorders—particularly stuttering—would be a plus.</span></p>
</div>
</div>
<div class="gmail-grand-item-offre">
<h2 class="gmail-h2-detail gmail-list-group-item-heading">Skills</h2>
<div class="gmail-list-group-item-text gmail-offre-detail-affiche-context">
<p> </p>
<p class="gmail-p1"><strong>Expected Skills</strong></p>
<p class="gmail-p2">The candidate should hold a master’s degree in computer
science, with strong skills in machine learning and deep learning, solid
proficiency in Python and frameworks such as PyTorch or TensorFlow, as
well as an interest in signal processing (audio/video) and, ideally, in
NLP. Autonomy, rigor, critical thinking, and analytical abilities are
essential, as well as good communication skills to thrive in a
multidisciplinary environment. An interest in phonetics, linguistics,
and speech disorders—particularly stuttering—will be a plus. The
candidate should also have the ability to work effectively in a
multidisciplinary team.</p>
</div>
</div>
<div class="gmail-grand-item-offre">
<h2 class="gmail-h2-detail gmail-list-group-item-heading">Benefits package</h2>
<div class="gmail-list-group-item-text gmail-offre-detail-affiche-context">
<ul><li>Restauration subventionnée</li><li>Transports publics remboursés partiellement</li><li>Congés: 7 semaines de congés annuels + 10 jours de RTT (base temps
plein) + possibilité d'autorisations d'absence exceptionnelle (ex :
enfants malades, déménagement)</li><li>Possibilité de télétravail (après 6 mois d'ancienneté) et aménagement du temps de travail</li><li>Équipements professionnels à disposition (visioconférence, prêts de matériels informatiques, etc.)</li><li>Prestations sociales, culturelles et sportives (Association de gestion des œuvres sociales d'Inria)</li><li>Accès à la formation professionnelle</li><li>Sécurité sociale</li></ul>
</div>
</div>
<div class="gmail-grand-item-offre">
<h2 class="gmail-h2-detail gmail-list-group-item-heading">Remuneration</h2>
<div class="gmail-list-group-item-text gmail-offre-detail-affiche-context">
<p>€2300 gross/month</p>
</div><br></div><div class="gmail-grand-item-offre">Job Details: <a href="https://jobs.inria.fr/public/classic/en/offres/2025-09498/topdf">https://jobs.inria.fr/public/classic/en/offres/2025-09498/topdf</a></div><br clear="all"></div><br><span class="gmail_signature_prefix">-- </span><br><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><span style="font-family:Helvetica,sans-serif;color:rgb(0,111,201)">Kind Regards,</span><span style="font-family:Calibri,sans-serif;color:black"><br></span><span style="font-family:Helvetica,sans-serif;color:rgb(0,111,201)">Dr. Shakeel A. Sheikh,</span></div><div dir="ltr"><span style="font-family:Helvetica,sans-serif;color:rgb(0,111,201)">Prof Slim Ouni</span></div><div><br></div></div></div>