As text-to-speech technologies achieve remarkable naturalness in read-aloud tasks, there is growing interest in multimodal synthesis of verbal and non-verbal communicative behaviour, such as spontaneous speech and associated body gestures. This paper presents a novel, unified architecture for jointly synthesising speech acoustics and skeleton-based 3D gesture motion from text, trained using optimal-transport conditional flow matching (OT-CFM). The proposed architecture is simpler than the previous state of the art, has a smaller memory footprint, and can capture the joint distribution of speech and gestures, generating both modalities together in one single process. The new training regime, meanwhile, enables better synthesis quality in much fewer steps (network evaluations) than before. Uni- and multimodal subjective tests demonstrate improved speech naturalness, gesture human-likeness, and cross-modal appropriateness compared to existing benchmarks.

This paper focuses on enhancing human-agent communication by integrating spatial context into virtual agents’ non-verbal behaviors, specifically gestures. Recent advances in co-speech gesture generation have primarily utilized data-driven methods, which create natural motion but limit the scope of gestures to those performed in a void. Our work aims to extend these methods by enabling generative models to incorporate scene information into speech-driven gesture synthesis. We introduce a novel synthetic gesture dataset tailored for this purpose. This development represents a critical step toward creating embodied conversational agents that interact more naturally with their environment and users.

Embodied conversational agents and social robots need to be able to generate spontaneous behavior in order to be believable in social interactions. We present a system that can generate spontaneous speech with supporting lip movements. The conversational TTS voice is trained on a podcast corpus that has been prosodically tagged (f0, speaking rate and energy) and transcribed (including tokens for breathing, fillers and laughter). We introduce a speech animation algorithm where articulatory effort can be adjusted. The speech animation is driven by time-stamped phonemes obtained from the internal alignment attention map of the TTS system, and we use prominence estimates from the synthesised speech waveform to modulate the lip- and jaw movements accordingly.

Diffusion models have experienced a surge of interest as highly expressive yet efficiently trainable probabilistic models. We show that these models are an excellent fit for synthesising human motion that co-occurs with audio, e.g., dancing and co-speech gesticulation, since motion is complex and highly ambiguous given audio, calling for a probabilistic description. Specifically, we adapt the DiffWave architecture to model 3D pose sequences, putting Conformers in place of dilated convolutions for improved modelling power. We also demonstrate control over motion style, using classifier-free guidance to adjust the strength of the stylistic expression. Experiments on gesture and dance generation confirm that the proposed method achieves top-of-the-line motion quality, with distinctive styles whose expression can be made more or less pronounced. We also synthesise path-driven locomotion using the same model architecture. Finally, we generalise the guidance procedure to obtain product-of-expert ensembles of diffusion models and demonstrate how these may be used for, e.g., style interpolation, a contribution we believe is of independent interest.

Text-to-speech and co-speech gesture synthesis have until now been treated as separate areas by two different research communities, and applications merely stack the two technologies using a simple system-level pipeline. This can lead to modeling inefficiencies and may introduce inconsistencies that limit the achievable naturalness. We propose to instead synthesize the two modalities in a single model, a new problem we call integrated speech and gesture synthesis (ISG). We also propose a set of models modified from state-of-the-art neural speech-synthesis engines to achieve this goal. We evaluate the models in three carefully-designed user studies, two of which evaluate the synthesized speech and gesture in isolation, plus a combined study that evaluates the models like they will be used in real-world applications – speech and gesture presented together. The results show that participants rate one of the proposed integrated synthesis models as being as good as the state-of-the-art pipeline system we compare against, in all three tests. The model is able to achieve this with faster synthesis time and greatly reduced parameter count compared to the pipeline system, illustrating some of the potential benefits of treating speech and gesture synthesis together as a single, unified problem.

Motion capture is a well-established technology for capturing actors' movements and performances within the entertainment industry. Many actors, however, witness the poor acting conditions associated with such recordings. Instead of detailed sets, costumes and props, they are forced to play in empty spaces wearing tight suits. Often, their co-actors will be imaginary, replaced by placeholder props, or they would be out of scale with their virtual counterparts. These problems do not only affect acting, they also cause an abundance of laborious post-processing clean-up work. To solve these challenges, we propose using a combination of virtual reality and motion capture technology to bring differently proportioned virtual characters into a shared collaborative virtual environment. A within-subjects user study with trained actors showed that our proposed platform enhances their feelings of body ownership and immersion. This in turn changed actors' performances which narrowed the gap between virtual performances and final intended animations.

Dance requires skillful composition of complex movements that follow rhythmic, tonal and timbral features of music. Formally, generating dance conditioned on a piece of music can be expressed as a problem of modelling a high-dimensional continuous motion signal, conditioned on an audio signal. In this work we make two contributions to tackle this problem. First, we present a novel probabilistic autoregressive architecture that models the distribution over future poses with a normalizing flow conditioned on previous poses as well as music context, using a multimodal transformer encoder. Second, we introduce the currently largest 3D dance-motion dataset, obtained with a variety of motion-capture technologies, and including both professional and casual dancers. Using this dataset, we compare our new model against two baselines, via objective metrics and a user study, and show that both the ability to model a probability distribution, as well as being able to attend over a large motion and music context are necessary to produce interesting, diverse, and realistic dance that matches the music.

During speech, people spontaneously gesticulate, which plays a key role in conveying information. Similarly, realistic co-speech gestures are crucial to enable natural and smooth interactions with social agents. Current end-to-end co-speech gesture generation systems use a single modality for representing speech: either audio or text. These systems are therefore confined to producing either acoustically-linked beat gestures or semantically-linked gesticulation (e.g., raising a hand when saying "high''): they cannot appropriately learn to generate both gesture types. We present a model designed to produce arbitrary beat and semantic gestures together. Our deep-learning based model takes both acoustic and semantic representations of speech as input, and generates gestures as a sequence of joint angle rotations as output. The resulting gestures can be applied to both virtual agents and humanoid robots. Subjective and objective evaluations confirm the success of our approach. The code and video are available at the project page svito-zar.github.io/gesticulator .

Embodied human communication encompasses both verbal (speech) and non-verbal information (e.g., gesture and head movements). Recent advances in machine learning have substantially improved the technologies for generating synthetic versions of both of these types of data: On the speech side, text-to-speech systems are now able to generate highly convincing, spontaneous-sounding speech using unscripted speech audio as the source material. On the motion side, probabilistic motion-generation methods can now synthesise vivid and lifelike speech-driven 3D gesticulation. In this paper, we put these two state-of-the-art technologies together in a coherent fashion for the first time. Concretely, we demonstrate a proof-of-concept system trained on a single-speaker audio and motion-capture dataset, that is able to generate both speech and full-body gestures together from text input. In contrast to previous approaches for joint speech-and-gesture generation, we generate full-body gestures from speech synthesis trained on recordings of spontaneous speech from the same person as the motion-capture data. We illustrate our results by visualising gesture spaces and textspeech-gesture alignments, and through a demonstration video.

Abstract Automatic synthesis of realistic gestures promises to transform the fields of animation, avatars and communicative agents. In off‐line applications, novel tools can alter the role of an animator to that of a director, who provides only high‐level input for the desired animation; a learned network then translates these instructions into an appropriate sequence of body poses. In interactive scenarios, systems for generating natural animations on the fly are key to achieving believable and relatable characters. In this paper we address some of the core issues towards these ends. By adapting a deep learning‐based motion synthesis method called MoGlow, we propose a new generative model for generating state‐of‐the‐art realistic speech‐driven gesticulation. Owing to the probabilistic nature of the approach, our model can produce a battery of different, yet plausible, gestures given the same input speech signal. Just like humans, this gives a rich natural variation of motion. We additionally demonstrate the ability to exert directorial control over the output style, such as gesture level, speed, symmetry and spacial extent. Such control can be leveraged to convey a desired character personality or mood. We achieve all this without any manual annotation of the data. User studies evaluating upper‐body gesticulation confirm that the generated motions are natural and well match the input speech. Our method scores above all prior systems and baselines on these measures, and comes close to the ratings of the original recorded motions. We furthermore find that we can accurately control gesticulation styles without unnecessarily compromising perceived naturalness. Finally, we also demonstrate an application of the same method to full‐body gesticulation, including the synthesis of stepping motion and stance.

Data-driven modelling and synthesis of motion is an active research area with applications that include animation, games, and social robotics. This paper introduces a new class of probabilistic, generative, and controllable motion-data models based on normalising flows. Models of this kind can describe highly complex distributions, yet can be trained efficiently using exact maximum likelihood, unlike GANs or VAEs. Our proposed model is autoregressive and uses LSTMs to enable arbitrarily long time-dependencies. Importantly, is is also causal, meaning that each pose in the output sequence is generated without access to poses or control inputs from future time steps; this absence of algorithmic latency is important for interactive applications with real-time motion control. The approach can in principle be applied to any type of motion since it does not make restrictive, task-specific assumptions regarding the motion or the character morphology. We evaluate the models on motion-capture datasets of human and quadruped locomotion. Objective and subjective results show that randomly-sampled motion from the proposed method outperforms task-agnostic baselines and attains a motion quality close to recorded motion capture.

Most of the motion re-targeting techniques are grounded on virtual character animation research, which means that they typically assume that the target embodiment has unconstrained joint angular velocities. However, because robots often do have such constraints, traditional re-targeting approaches can originate irregular delays in the robot motion. With the goal of ensuring synchronization between verbal and nonverbal behavior, this paper proposes an optimization framework for processing re-targeted motion sequences that addresses constraints such as joint angle and angular velocities. The proposed framework was evaluated on a humanoid robot using both objective and subjective metrics. While the analysis of the joint motion trajectories provides evidence that our framework successfully performs the desired modifications to ensure verbal and nonverbal behavior synchronization, results from a perceptual study showed that participants found the robot motion generated by our method more natural, elegant and lifelike than a control condition.

This paper presents a unique dataset consisting of 20 recordings of the same musical piece, conducted with 4 different musical intentions in mind. The upper body and baton motion of a professional conductor was recorded, as well as the sound of each instrument in a professional string quartet following the conductor. The dataset is made available for benchmarking of motion recognition algorithms. An HMM-based emotion intent classification method is trained with subsets of the data, and classification of other subsets of the data show firstly that the motion of the baton communicates energetic intention to a high degree, secondly, that the conductor’s torso, head and other arm conveys calm intention to a high degree, and thirdly, that positive vs negative sentiments are communicated to a high degree through other channels than the body and baton motion – most probably, through facial expression and muscle tension conveyed through articulated hand and finger motion. The long-term goal of this work is to develop a computer model of the entire conductor-orchestra communication process; the studies presented here indicate that computer modeling of the conductor-orchestra communication is feasible.

Unlike their human counterparts, artificial agents such as robots and game characters may be deployed with a large variety of face and body configurations. Some have articulated bodies but lack facial features, and others may be talking heads ending at the neck. Generally, they have many fewer degrees of freedom than humans through which they must express themselves, and there will inevitably be a filtering effect when mapping human motion onto the agent. In this article, we investigate filtering effects on three types of embodiments: (a) an agent with a body but no facial features, (b) an agent with a head only, and (c) an agent with a body and a face. We performed a full performance capture of a mime actor enacting short interactions varying the non-verbal expression along five dimensions (e.g., level of frustration and level of certainty) for each of the three embodiments. We performed a crowd-sourced evaluation experiment comparing the video of the actor to the video of an animated robot for the different embodiments and dimensions. Our findings suggest that the face is especially important to pinpoint emotional reactions but is also most volatile to filtering effects. The body motion, on the other hand, had more diverse interpretations but tended to preserve the interpretation after mapping and thus proved to be more resilient to filtering.

Passive optical motion capture is one of the predominant technologies for capturing high fidelity human skeletal motion, and is a workhorse in a large number of areas such as bio-mechanics, film and video games. While most state-of-the-art systems can automatically identify and track markers on the larger parts of the human body, the markers attached to fingers provide unique challenges and usually require extensive manual cleanup. In this work we present a robust online method for identification and tracking of passive motion capture markers attached to the fingers of the hands. The method is especially suited for large capture volumes and sparse marker sets of 3 to 10 markers per hand. Once trained, our system can automatically initialize and track the markers, and the subject may exit and enter the capture volume at will. By using multiple assignment hypotheses and soft decisions, it can robustly recover from a difficult situation with many simultaneous occlusions and false observations (ghost markers). We evaluate the method on a collection of sparse marker sets commonly used in industry and in the research community. We also compare the results with two of the most widely used motion capture platforms: Motion Analysis Cortex and Vicon Blade. The results show that our method is better at attaining correct marker labels and is especially beneficial for real-time applications.

In this paper we study the production and perception of speech in diverse conditions for the purposes of accurate, flexible and highly intelligible talking face animation. We recorded audio, video and facial motion capture data of a talker uttering a set of 180 short sentences, under three conditions: normal speech (in quiet), Lombard speech (in noise), and whispering. We then produced an animated 3D avatar with similar shape and appearance as the original talker and used an error minimization procedure to drive the animated version of the talker in a way that matched the original performance as closely as possible. In a perceptual intelligibility study with degraded audio we then compared the animated talker against the real talker and the audio alone, in terms of audio-visual word recognition rate across the three different production conditions. We found that the visual intelligibility of the animated talker was on par with the real talker for the Lombard and whisper conditions. In addition we created two incongruent conditions where normal speech audio was paired with animated Lombard speech or whispering. When compared to the congruent normal speech condition, Lombard animation yields a significant increase in intelligibility, despite the AV-incongruence. In a separate evaluation, we gathered subjective opinions on the different animations, and found that some degree of incongruence was generally accepted.