# Co-Speech Gesture Synthesis using Discrete Gesture Token Learning

arXiv: https://arxiv.org/abs/2303.12822 · HTML: https://mlsanity.com/paper/2303.12822/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 34 references that are in the map.
- First posted on arXiv 4 Mar 2023.

## Builds on (most cited first, 27 of 27)

- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [Zero-Shot Text-to-Image Generation](https://mlsanity.com/paper/2102.12092/) (2021)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [Expressive Body Capture: 3D Hands, Face, and Body from a Single Image](https://mlsanity.com/paper/1904.05866/) (2019)
- [Enriching Word Vectors with Subword Information](https://mlsanity.com/paper/1607.04606/) (2016)
- [An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling](https://mlsanity.com/paper/1803.01271/) (2018)
- [Generating Diverse High-Fidelity Images with VQ-VAE-2](https://mlsanity.com/paper/1906.00446/) (2019)
- [On the Continuity of Rotation Representations in Neural Networks](https://mlsanity.com/paper/1812.07035/) (2018)
- [OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields](https://mlsanity.com/paper/1812.08008/) (2018)
- [Autoregressive Image Generation using Residual Quantization](https://mlsanity.com/paper/2203.01941/) (2022)
- [VideoGPT: Video Generation using VQ-VAE and Transformers](https://mlsanity.com/paper/2104.10157/) (2021)
- [Jukebox: A Generative Model for Music](https://mlsanity.com/paper/2005.00341/) (2020)
- [Language2Pose: Natural Language Grounded Pose Forecasting](https://mlsanity.com/paper/1907.01108/) (2019)
- [Speech Gesture Generation from the Trimodal Context of Text, Audio, and Speaker Identity](https://mlsanity.com/paper/2009.02119/) (2020)
- [Learning Individual Styles of Conversational Gesture](https://mlsanity.com/paper/1906.04160/) (2019)
- [Robots Learn Social Skills: End-to-End Learning of Co-Speech Gesture Generation for Humanoid Robots](https://mlsanity.com/paper/1810.12541/) (2018)
- [Audio2Gestures: Generating Diverse Gestures from Speech Audio with Conditional Variational Autoencoders](https://mlsanity.com/paper/2108.06720/) (2021)
- [Learning Hierarchical Cross-Modal Association for Co-Speech Gesture Generation](https://mlsanity.com/paper/2203.13161/) (2022)
- [Text2Gestures: A Transformer-Based Network for Generating Emotive Body Gestures for Virtual Agents](https://mlsanity.com/paper/2101.11101/) (2021)
- [Gesticulator: A framework for semantically-aware speech-driven gesture generation](https://mlsanity.com/paper/2001.09326/) (2020)
- [ZeroEGGS: Zero-shot Example-based Gesture Generation from Speech](https://mlsanity.com/paper/2209.07556/) (2022)
- [Speech2AffectiveGestures: Synthesizing Co-Speech Gestures with Generative Adversarial Affective Expression Learning](https://mlsanity.com/paper/2108.00262/) (2021)
- [Speech Drives Templates: Co-Speech Gesture Synthesis with Learned Templates](https://mlsanity.com/paper/2108.08020/) (2021)
- [A large, crowdsourced evaluation of gesture generation systems on common data: The GENEA Challenge 2020](https://mlsanity.com/paper/2102.11617/) (2021)
- [Multi-Instrumentalist Net: Unsupervised Generation of Music from Body Movements](https://mlsanity.com/paper/2012.03478/) (2020)
- [HEMVIP: Human Evaluation of Multiple Videos in Parallel](https://mlsanity.com/paper/2101.11898/) (2021)
- [SGToolkit: An Interactive Gesture Authoring Toolkit for Embodied Conversational Agents](https://mlsanity.com/paper/2108.04636/) (2021)

## Built on by (most cited first, 9 of 9)

- [EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Expressive Masked Audio Gesture Modeling](https://mlsanity.com/paper/2401.00374/) (2023)
- [MambaTalk: Efficient Holistic Gesture Synthesis with Selective State Space Models](https://mlsanity.com/paper/2403.09471/) (2024)
- [Emotional Speech-driven 3D Body Animation via Disentangled Latent Diffusion](https://mlsanity.com/paper/2312.04466/) (2023)
- [RapVerse: Coherent Vocals and Whole-Body Motions Generations from Text](https://mlsanity.com/paper/2405.20336/) (2024)
- [Fake it to make it: Using synthetic data to remedy the data shortage in joint multimodal speech-and-gesture synthesis](https://mlsanity.com/paper/2404.19622/) (2024)
- [C2G2: Controllable Co-speech Gesture Generation with Latent Diffusion Model](https://mlsanity.com/paper/2308.15016/) (2023)
- [NVP-HRI: Zero Shot Natural Voice and Posture-based Human-Robot Interaction via Large Language Model](https://mlsanity.com/paper/2503.09335/) (2025)
- [Generating Natural and Expressive Robot Gestures through Iterative Reinforcement Learning with Human Feedback using LLMs](https://mlsanity.com/paper/2606.18747/) (2026)
- [SocialHumanoid: Towards Expressive Humanoid Behavior via One-Step Co-Speech Motion Generation](https://mlsanity.com/paper/2609.33311/) (2026)

## Abstract (arXiv)

Synthesizing realistic co-speech gestures is an important and yet unsolved problem for creating believable motions that can drive a humanoid robot to interact and communicate with human users. Such capability will improve the impressions of the robots by human users and will find applications in education, training, and medical services. One challenge in learning the co-speech gesture model is that there may be multiple viable gesture motions for the same speech utterance. The deterministic regression methods can not resolve the conflicting samples and may produce over-smoothed or damped motions. We proposed a two-stage model to address this uncertainty issue in gesture synthesis by modeling the gesture segments as discrete latent codes. Our method utilizes RQ-VAE in the first stage to learn a discrete codebook consisting of gesture tokens from training data. In the second stage, a two-level autoregressive transformer model is used to learn the prior distribution of residual codes conditioned on input speech context. Since the inference is formulated as token sampling, multiple gesture sequences could be generated given the same speech input using top-k sampling. The quantitative results and the user study showed the proposed method outperforms the previous methods and is able to generate realistic and diverse gesture motions.