Catherine Pelachaud

Mireille Fares, Michele Grimaldi, Catherine Pelachaud et al.

Modeling virtual agents with behavior style is one factor for personalizing human agent interaction. We propose an efficient yet effective machine learning approach to synthesize gestures driven by prosodic features and text in the style of different speakers including those unseen during training. Our model performs zero shot multimodal style transfer driven by multimodal data from the PATS database containing videos of various speakers. We view style as being pervasive while speaking, it colors the communicative behaviors expressivity while speech content is carried by multimodal signals and text. This disentanglement scheme of content and style allows us to directly infer the style embedding even of speaker whose data are not part of the training phase, without requiring any further training or fine tuning. The first goal of our model is to generate the gestures of a source speaker based on the content of two audio and text modalities. The second goal is to condition the source speaker predicted gestures on the multimodal behavior style embedding of a target speaker. The third goal is to allow zero shot style transfer of speakers unseen during training without retraining the model. Our system consists of: (1) a speaker style encoder network that learns to generate a fixed dimensional speaker embedding style from a target speaker multimodal data and (2) a sequence to sequence synthesis network that synthesizes gestures based on the content of the input modalities of a source speaker and conditioned on the speaker style embedding. We evaluate that our model can synthesize gestures of a source speaker and transfer the knowledge of target speaker style variability to the gesture generation task in a zero shot setup. We convert the 2D gestures to 3D poses and produce 3D animations. We conduct objective and subjective evaluations to validate our approach and compare it with a baseline.

Mireille Fares, Catherine Pelachaud, Nicolas Obin

This paper addresses the challenge of transferring the behavior expressivity style of a virtual agent to another one while preserving behaviors shape as they carry communicative meaning. Behavior expressivity style is viewed here as the qualitative properties of behaviors. We propose TranSTYLer, a multimodal transformer based model that synthesizes the multimodal behaviors of a source speaker with the style of a target speaker. We assume that behavior expressivity style is encoded across various modalities of communication, including text, speech, body gestures, and facial expressions. The model employs a style and content disentanglement schema to ensure that the transferred style does not interfere with the meaning conveyed by the source behaviors. Our approach eliminates the need for style labels and allows the generalization to styles that have not been seen during the training phase. We train our model on the PATS corpus, which we extended to include dialog acts and 2D facial landmarks. Objective and subjective evaluations show that our model outperforms state of the art models in style transfer for both seen and unseen styles during training. To tackle the issues of style and content leakage that may arise, we propose a methodology to assess the degree to which behavior and gestures associated with the target style are successfully transferred, while ensuring the preservation of the ones related to the source content.

Fajrian Yunus, Chloé Clavel, Catherine Pelachaud

Image schema is a recurrent pattern of reasoning where one entity is mapped into another. Image schema is similar to conceptual metaphor and is also related to metaphoric gesture. Our main goal is to generate metaphoric gestures for an Embodied Conversational Agent. We propose a technique to learn the vector representation of image schemas. As far as we are aware of, this is the first work which addresses that problem. Our technique uses Ravenet et al's algorithm which we use to compute the image schemas from the text input and also BERT and SenseBERT which we use as the base word embedding technique to calculate the final vector representation of the image schema. Our representation learning technique works by clustering: word embedding vectors which belong to the same image schema should be relatively closer to each other, and thus form a cluster. With the image schemas representable as vectors, it also becomes possible to have a notion that some image schemas are closer or more similar to each other than to the others because the distance between the vectors is a proxy of the dissimilarity between the corresponding image schemas. Therefore, after obtaining the vector representation of the image schemas, we calculate the distances between those vectors. Based on these, we create visualizations to illustrate the relative distances between the different image schemas.

Mireille Fares, Catherine Pelachaud, Nicolas Obin

Image Schemas are repetitive cognitive patterns that influence the way we conceptualize and reason about various concepts present in speech. These patterns are deeply embedded within our cognitive processes and are reflected in our bodily expressions including gestures. Particularly, metaphoric gestures possess essential characteristics and semantic meanings that align with Image Schemas, to visually represent abstract concepts. The shape and form of gestures can convey abstract concepts, such as extending the forearm and hand or tracing a line with hand movements to visually represent the image schema of PATH. Previous behavior generation models have primarily focused on utilizing speech (acoustic features and text) to drive the generation model of virtual agents. They have not considered key semantic information as those carried by Image Schemas to effectively generate metaphoric gestures. To address this limitation, we introduce META4, a deep learning approach that generates metaphoric gestures from both speech and Image Schemas. Our approach has two primary goals: computing Image Schemas from input text to capture the underlying semantic and metaphorical meaning, and generating metaphoric gestures driven by speech and the computed image schemas. Our approach is the first method for generating speech driven metaphoric gestures while leveraging the potential of Image Schemas. We demonstrate the effectiveness of our approach and highlight the importance of both speech and image schemas in modeling metaphoric gestures.

Lucie Galland, Catherine Pelachaud, Florian Pecune

Motivational Interviewing (MI) is an approach to therapy that emphasizes collaboration and encourages behavioral change. To evaluate the quality of an MI conversation, client utterances can be classified using the MISC code as either change talk, sustain talk, or follow/neutral talk. The proportion of change talk in a MI conversation is positively correlated with therapy outcomes, making accurate classification of client utterances essential. In this paper, we present a classifier that accurately distinguishes between the three MISC classes (change talk, sustain talk, and follow/neutral talk) leveraging multimodal features such as text, prosody, facial expressivity, and body expressivity. To train our model, we perform annotations on the publicly available AnnoMI dataset to collect multimodal information, including text, audio, facial expressivity, and body expressivity. Furthermore, we identify the most important modalities in the decision-making process, providing valuable insights into the interplay of different modalities during a MI conversation.

Yann Munro, Camilo Sarmiento, Isabelle Bloch et al.

An abstract argumentation framework is a commonly used formalism to provide a static representation of a dialogue. However, the order of enunciation of the arguments in an argumentative dialogue is very important and can affect the outcome of this dialogue. In this paper, we propose a new framework for modelling abstract argumentation graphs, a model that incorporates the order of enunciation of arguments. By taking this order into account, we have the means to deduce a unique outcome for each dialogue, called an extension. We also establish several properties, such as termination and correctness, and discuss two notions of completeness. In particular, we propose a modification of the previous transformation based on a "last enunciated last updated" strategy, which verifies the second form of completeness.

Takeshi Saga, Catherine Pelachaud

Turn-taking management is crucial for any social interaction. Still, it is challenging to model human-machine interaction due to the complexity of the social context and its multimodal nature. Unlike conventional systems based on silence duration, previous existing voice activity projection (VAP) models successfully utilized a unified representation of turn-taking behaviors as prediction targets, which improved turn-taking prediction performance. Recently, a multimodal VAP model outperformed the previous state-of-the-art model by a significant margin. In this paper, we propose a multimodal model enhanced with pre-trained audio and face encoders to improve performance by capturing subtle expressions. Our model performed competitively, and in some cases, even better than state-of-the-art models on turn-taking metrics. All the source codes and pretrained models are available at https://github.com/sagatake/VAPwithAudioFaceEncoders.

Téo Guichoux, Laure Soulier, Nicolas Obin et al.

Co-speech gestures are fundamental for communication. The advent of recent deep learning techniques has facilitated the creation of lifelike, synchronous co-speech gestures for Embodied Conversational Agents. "In-the-wild" datasets, aggregating video content from platforms like YouTube via human pose detection technologies, provide a feasible solution by offering 2D skeletal sequences aligned with speech. Concurrent developments in lifting models enable the conversion of these 2D sequences into 3D gesture databases. However, it is important to note that the 3D poses estimated from the 2D extracted poses are, in essence, approximations of the ground-truth, which remains in the 2D domain. This distinction raises questions about the impact of gesture representation dimensionality on the quality of generated motions - a topic that, to our knowledge, remains largely unexplored. Our study examines the effect of using either 2D or 3D joint coordinates as training data on the performance of speech-to-gesture deep generative models. We employ a lifting model for converting generated 2D pose sequences into 3D and assess how gestures created directly in 3D stack up against those initially generated in 2D and then converted to 3D. We perform an objective evaluation using widely used metrics in the gesture generation field as well as a user study to qualitatively evaluate the different approaches.

Lucrezia Tosato, Victor Fortier, Isabelle Bloch et al.

Studies in human human interaction have introduced the concept of F formation to describe the spatial arrangement of participants during social interactions. This paper has two objectives. It aims at detecting F formations in video sequences and predicting the next speaker in a group conversation. The proposed approach exploits time information and human multimodal signals in video sequences. In particular, we rely on measuring the engagement level of people as a feature of group belonging. Our approach makes use of a recursive neural network, the Long Short Term Memory (LSTM), to predict who will take the speaker's turn in a conversation group. Experiments on the MatchNMingle dataset led to 85% true positives in group detection and 98% accuracy in predicting the next speaker.

Anh Ngo, Nicolas Rollet, Catherine Pelachaud et al.

Maintaining mutual understanding is a key component in human-human conversation to avoid conversation breakdowns, in which repair, particularly Other-Initiated Repair (OIR, when one speaker signals trouble and prompts the other to resolve), plays a vital role. However, Conversational Agents (CAs) still fail to recognize user repair initiation, leading to breakdowns or disengagement. This work proposes a multimodal model to automatically detect repair initiation in Dutch dialogues by integrating linguistic and prosodic features grounded in Conversation Analysis. The results show that prosodic cues complement linguistic features and significantly improve the results of pretrained text and audio embeddings, offering insights into how different features interact. Future directions include incorporating visual cues, exploring multilingual and cross-context corpora to assess the robustness and generalizability.

Téo Guichoux, Théodor Lemerle, Shivam Mehta et al.

Human communication is multimodal, with speech and gestures tightly coupled, yet most computational methods for generating speech and gestures synthesize them sequentially, weakening synchrony and prosody alignment. We introduce Gelina, a unified framework that jointly synthesizes speech and co-speech gestures from text using interleaved token sequences in a discrete autoregressive backbone, with modality-specific decoders. Gelina supports multi-speaker and multi-style cloning and enables gesture-only synthesis from speech inputs. Subjective and objective evaluations demonstrate competitive speech quality and improved gesture generation over unimodal baselines.

Lucie Galland, Catherine Pelachaud, Florian Pecune

In this work, we propose a novel framework that integrates large language models (LLMs) with an RL-based dialogue manager for open-ended dialogue with a specific goal. By leveraging hierarchical reinforcement learning to model the structured phases of dialogue and employ meta-learning to enhance adaptability across diverse user profiles, our approach enhances adaptability and efficiency, enabling the system to learn from limited data, transition fluidly between dialogue phases, and personalize responses to heterogeneous patient needs. We apply our framework to Motivational Interviews, aiming to foster behavior change, and demonstrate that the proposed dialogue manager outperforms a state-of-the-art LLM baseline in terms of reward, showing a potential benefit of conditioning LLMs to create open-ended dialogue systems with specific goals.

Lucie Galland, Catherine Pelachaud, Florian Pecune

The study of multimodal interaction in therapy can yield a comprehensive understanding of therapist and patient behavior that can be used to develop a multimodal virtual agent supporting therapy. This investigation aims to uncover how therapists skillfully blend therapy's task goal (employing classical steps of Motivational Interviewing) with the social goal (building a trusting relationship and expressing empathy). Furthermore, we seek to categorize patients into various ``types'' requiring tailored therapeutic approaches. To this intent, we present multimodal annotations of a corpus consisting of simulated motivational interviewing conversations, wherein actors portray the roles of patients and therapists. We introduce EMMI, composed of two publicly available MI corpora, AnnoMI and the Motivational Interviewing Dataset, for which we add multimodal annotations. We analyze these annotations to characterize functional behavior for developing a virtual agent performing motivational interviews emphasizing social and empathic behaviors. Our analysis found three clusters of patients expressing significant differences in behavior and adaptation of the therapist's behavior to those types. This shows the importance of a therapist being able to adapt their behavior depending on the current situation within the dialog and the type of user.

Teo Guichoux, Laure Soulier, Nicolas Obin et al.

Co-speech gestures play a crucial role in the interactions between humans and embodied conversational agents (ECA). Recent deep learning methods enable the generation of realistic, natural co-speech gestures synchronized with speech, but such approaches require large amounts of training data. "In-the-wild" datasets, which compile videos from sources such as YouTube through human pose detection models, offer a solution by providing 2D skeleton sequences that are paired with speech. Concurrently, innovative lifting models have emerged, capable of transforming these 2D pose sequences into their 3D counterparts, leading to large and diverse datasets of 3D gestures. However, the derived 3D pose estimation is essentially a pseudo-ground truth, with the actual ground truth being the 2D motion data. This distinction raises questions about the impact of gesture representation dimensionality on the quality of generated motions, a topic that, to our knowledge, remains largely unexplored. In this work, we evaluate the impact of the dimensionality of the training data, 2D or 3D joint coordinates, on the performance of a multimodal speech-to-gesture deep generative model. We use a lifting model to convert 2D-generated sequences of body pose to 3D. Then, we compare the sequence of gestures generated directly in 3D to the gestures generated in 2D and lifted to 3D as post-processing.

Mireille Fares, Catherine Pelachaud, Nicolas Obin

In this study, we address the importance of modeling behavior style in virtual agents for personalized human-agent interaction. We propose a machine learning approach to synthesize gestures, driven by prosodic features and text, in the style of different speakers, even those unseen during training. Our model incorporates zero-shot multimodal style transfer using multimodal data from the PATS database, which contains videos of diverse speakers. We recognize style as a pervasive element during speech, influencing the expressivity of communicative behaviors, while content is conveyed through multimodal signals and text. By disentangling content and style, we directly infer the style embedding, even for speakers not included in the training phase, without the need for additional training or fine-tuning. Objective and subjective evaluations are conducted to validate our approach and compare it against two baseline methods.

Jieyeon Woo, Mireille Fares, Catherine Pelachaud et al.

Socially Interactive Agents (SIAs) are physical or virtual embodied agents that display similar behavior as human multimodal behavior. Modeling SIAs' non-verbal behavior, such as speech and facial gestures, has always been a challenging task, given that a SIA can take the role of a speaker or a listener. A SIA must emit appropriate behavior adapted to its own speech, its previous behaviors (intra-personal), and the User's behaviors (inter-personal) for both roles. We propose AMII, a novel approach to synthesize adaptive facial gestures for SIAs while interacting with Users and acting interchangeably as a speaker or as a listener. AMII is characterized by modality memory encoding schema - where modality corresponds to either speech or facial gestures - and makes use of attention mechanisms to capture the intra-personal and inter-personal relationships. We validate our approach by conducting objective evaluations and comparing it with the state-of-the-art approaches.

Fajrian Yunus, Chloé Clavel, Catherine Pelachaud

Communicative gestures and speech acoustic are tightly linked. Our objective is to predict the timing of gestures according to the acoustic. That is, we want to predict when a certain gesture occurs. We develop a model based on a recurrent neural network with attention mechanism. The model is trained on a corpus of natural dyadic interaction where the speech acoustic and the gesture phases and types have been annotated. The input of the model is a sequence of speech acoustic and the output is a sequence of gesture classes. The classes we are using for the model output is based on a combination of gesture phases and gesture types. We use a sequence comparison technique to evaluate the model performance. We find that the model can predict better certain gesture classes than others. We also perform ablation studies which reveal that fundamental frequency is a relevant feature for gesture prediction task. In another sub-experiment, we find that including eyebrow movements as acting as beat gesture improves the performance. Besides, we also find that a model trained on the data of one given speaker also works for the other speaker of the same conversation. We also perform a subjective experiment to measure how respondents judge the naturalness, the time consistency, and the semantic consistency of the generated gesture timing of a virtual agent. Our respondents rate the output of our model favorably.

Reshmashree B. Kantharaju, Dominic De Franco, Alison Pease et al.

Virtual humans need to be persuasive in order to promote behaviour change in human users. While several studies have focused on understanding the numerous aspects that influence the degree of persuasion, most of them are limited to dyadic interactions. In this paper, we present an evaluation study focused on understanding the effects of multiple agents on user's persuasion. Along with gender and status (authoritative & peer), we also look at type of focus employed by the agent i.e., user-directed where the agent aims to persuade by addressing the user directly and vicarious where the agent aims to persuade the user, who is an observer, indirectly by engaging another agent in the discussion. Participants were randomly assigned to one of the 12 conditions and presented with a persuasive message by one or several virtual agents. A questionnaire was used to measure perceived interpersonal attitude, credibility and persuasion. Results indicate that credibility positively affects persuasion. In general, multiple agent setting, irrespective of the focus, was more persuasive than single agent setting. Although, participants favored user-directed setting and reported it to be persuasive and had an increased level of trust in the agents, the actual change in persuasion score reflects that vicarious setting was the most effective in inducing behaviour change. In addition to this, the study also revealed that authoritative agents were the most persuasive.

Kevin Sanlaville, Gérard Assayag, Frédéric Bevilacqua et al.

In a Human-Computer Interaction context, we aim to elaborate an adaptive and generic interaction model in two different use cases: Embodied Conversational Agents and Creative Musical Agents for musical improvisation. To reach this goal, we'll try to use the concepts of adaptation and synchronization to enhance the interactive abilities of our agents and guide the development of our interaction model, and will try to make synchrony emerge from non-verbal dimensions of interaction.

Nicolas Sabouret, Hazaël Jones, Magalie Ochs et al.

The use of virtual agents in social coaching has increased rapidly in the last decade. In order to train the user in different situations than can occur in real life, the virtual agent should be able to express different social attitudes. In this paper, we propose a model of social attitudes that enables a virtual agent to reason on the appropriate social attitude to express during the interaction with a user given the course of the interaction, but also the emotions, mood and personality of the agent. Moreover, the model enables the virtual agent to display its social attitude through its non-verbal behaviour. The proposed model has been developed in the context of job interview simulation. The methodology used to develop such a model combined a theoretical and an empirical approach. Indeed, the model is based both on the literature in Human and Social Sciences on social attitudes but also on the analysis of an audiovisual corpus of job interviews and on post-hoc interviews with the recruiters on their expressed attitudes during the job interview.