Audio-Visual Digital Humans
We have been pursuing research on a number of aspects related to audio-visual digital humans. On this page we present the works that we have been involved with
1) Wav2Lip: Accurately Lip-syncing Videos In The Wild (ACM Multimedia 2020)
In this work, we investigate the problem of lip-syncing a talking face video of an arbitrary identity to match a target speech segment. Current works excel at producing accurate lip movements on a static image or on videos of specific people seen during the training phase. However, they fail to accurately morph the actual lip movements of arbitrary identities in dynamic, unconstrained talking face videos, resulting in significant parts of the video being out-of-sync with the newly chosen audio. We identify key reasons pertaining to this and hence resolve them by learning from a powerful lip-sync discriminator. Next, we propose new, rigorous evaluation benchmarks and metrics to specifically measure the accuracy of lip synchronization in unconstrained videos. Extensive quantitative and human evaluations on our challenging benchmarks show that the lip-sync accuracy of the videos generated using our Wav2Lip model is almost as good as real synced videos. We clearly demonstrate the substantial impact of our Wav2Lip model in our publicly available demo video. We also open-source our code, models, and evaluation benchmarks to promote future research efforts in this space.
2) Lip2Wav: Audio generation based on lip movements (CVPR 2020)
Humans involuntarily tend to infer parts of the conversation from lip movements when the speech is absent or corrupted by external noise. In this work, we explore the task of lip to speech synthesis, i.e., learning to generate natural speech given only the lip movements of a speaker. Acknowledging the importance of contextual and speaker-specific cues for accurate lip reading, we take a different path from existing works. We focus on learning accurate lip sequences to speech mappings for individual speakers in unconstrained, large vocabulary settings. To this end, we collect and release a large-scale benchmark dataset, the first of its kind, specifically to train and evaluate the single-speaker lip to speech task in natural settings. We propose an approach to achieve accurate, natural lip to speech synthesis in such unconstrained scenarios for the first time. Extensive evaluation using quantitative, qualitative metrics and human evaluation shows that our method is almost twice as intelligible as previous works in this space.
3) Towards Automatic Face-to-Face Translation (ACM Multimedia 2019)
In light of the recent breakthroughs in automatic machine translation systems, we propose a novel approach of what we term as “Face-to-Face Translation”. As today’s digital communication becomes increasingly visual, we argue that there is the need for systems that can automatically translate a video of a person speaking in language A into a target language B with realistic lip synchronization. In this work, we create an automatic pipeline for this problem and demonstrate its impact in multiple real-world applications. First, we build a working speech-to-speech translation system by bringing together multiple existing modules from speech and language. We then move towards “Face-to-Face Translation” by incorporating a novel visual module, LipGAN for generating realistic talking faces from the translated audio. Quantitative evaluation of LipGAN on the standard LRW test set, shows that it significantly outperforms existing approaches across all standard metrics. We also subject our Face-to-Face Translation pipeline, to multiple human evaluations and show that it can significantly improve the overall user experience for consuming and interacting with multimodal content across languages.
4) Visual Speech Enhancement Without a Real Visual Stream (WACV 2021)
In this work, we re-think the task of speech enhancement in unconstrained real-world environments. Current state-of-the-art methods use only the audio stream and are limited in their performance in a wide range of real-world noises. Recent works using lip movements as additional cues improve the quality of generated speech over “audio-only” methods. But, these methods cannot be used for several applications where the visual stream is unreliable or completely absent. We propose a new paradigm for speech enhancement by exploiting recent breakthroughs in speech-driven lip synthesis. Using one such model as a teacher network, we train a robust student network to produce accurate lip movements that mask away the noise, thus acting as a “visual noise filter”. The intelligibility of the speech enhanced by our pseudo-lip approach is almost close (< 3\% difference) to the case of using real lips. This implies that we can exploit the advantages of using lip movements even in the absence of a real video stream. We rigorously evaluate our model using quantitative metrics as well as qualitative human evaluations. Additional ablation studies and a demo video in the supplementary material containing qualitative comparisons and results clearly illustrate the effectiveness of our approach.
5) Extreme-scale Talking-Face Video Upsampling with Audio-Visual Priors (ACM Multimedia 2022)
In this paper, we explore an interesting question of what can be obtained from an 8×8 pixel video sequence. Surprisingly, it turns out to be quite a lot. We show that when we process this 8×8 video with the right set of audio and image priors, we can obtain a full-length, 256×256 video. We achieve this 32× scaling of an extremely low-resolution input using our novel audio-visual upsampling network. The audio prior helps to recover the elemental facial details and precise lip shapes and a single high-resolution target identity image prior provides us with rich appearance details. Our approach is an end-to-end multi-stage framework. The first stage produces a coarse intermediate output video that can be then used to animate single target identity image and generate realistic, accurate and high-quality outputs. Our approach is simple and performs exceedingly well (an 8× improvement in FID score) compared to previous super-resolution methods. We also extend our model to talking-face video compression, and show that we obtain a 3.5× improvement in terms of bits/pixel over the previous state-of-the-art. The results from our network are thoroughly analyzed through extensive ablation experiments (in the paper and supplementary material). We also provide the demo video along with code and models on our website
Video and demo coming soon
6) Personalized One-Shot Lipreading for an ALS Patient (BMVC 2021)
Lipreading or visually recognizing speech from the mouth movements of a speaker is a challenging and mentally taxing task. Unfortunately, multiple medical conditions force people to depend on this skill in their day-to-day lives for essential communication. Patients suffering from Amyotrophic Lateral Sclerosis (ALS) often lose muscle control, consequently their ability to generate speech and communicate via lip movements. Existing large datasets do not focus on medical patients or curate personalized vocabulary relevant to an individual. Collecting a large-scale dataset of a patient, needed to train mod-ern data-hungry deep learning models is, however, extremely challenging. In this work, we propose a personalized network to lipread an ALS patient using only one-shot examples. We depend on synthetically generated lip movements to augment the one-shot scenario. A Variational Encoder based domain adaptation technique is used to bridge the real-synthetic domain gap. Our approach significantly improves and achieves high top-5accuracy with 83.2% accuracy compared to 62.6% achieved by comparable methods for the patient. Apart from evaluating our approach on the ALS patient, we also extend it to people with hearing impairment relying extensively on lip movements to communicate.
7) FaceOff: A Video-to-Video Face Swapping System (WACV 2023)
Doubles play an indispensable role in the movie industry. They take the place of the actors in dangerous stunt scenes or in scenes where the same actor plays multiple characters. The double’s face is later replaced with the actor’s face and expressions manually using expensive CGI technology, costing millions of dollars and taking months to complete. An automated, inexpensive, and fast way can be to use face-swapping techniques that aim to swap an identity from a source face video (or an image) to a target face video. However, such methods can not preserve the source expressions of the actor important for the scene’s context. % essential for the scene. % that are essential in cinemas. To tackle this challenge, we introduce video-to-video (V2V) face-swapping, a novel task of face-swapping that can preserve (1) the identity and expressions of the source (actor) face video and (2) the background and pose of the target (double) video. We propose FaceOff, a V2V face-swapping system that operates by learning a robust blending operation to merge two face videos following the constraints above. It first reduces the videos to a quantized latent space and then blends them in the reduced space. FaceOff is trained in a self-supervised manner and robustly tackles the non-trivial challenges of V2V face-swapping. As shown in the experimental section, FaceOff significantly outperforms alternate approaches qualitatively and quantitatively.
8) Towards MOOCs for Lipreading: Using Synthetic Talking Heads to Train Humans in Lipreading at Scale (WACV 2023)
Many people with some form of hearing loss consider lipreading as their primary mode of day-to-day communication. However, finding resources to learn or improve one’s lipreading skills can be challenging. This is further exacerbated in COVID$19$ pandemic due to restrictions on direct interactions with peers and speech therapists. Today, online MOOCs platforms like Coursera and Udemy have become the most effective form of training for many kinds of skill development. However, online lipreading resources are scarce as creating such resources is an extensive process needing months of manual effort to record hired actors. Because of the manual pipeline, such platforms are also limited in the vocabulary, supported languages, accents, and speakers, and have a high usage cost. In this work, we investigate the possibility of replacing real human talking videos with synthetically generated videos. Synthetic data can be used to easily incorporate larger vocabularies, variations in accent, and even local languages, and many speakers. We propose an end-to-end automated pipeline to develop such a platform using state-of-the-art talking heading video generator networks, text-to-speech models, and computer vision techniques. We then perform an extensive human evaluation using carefully thought out lipreading exercises to validate the quality of our designed platform against the existing lipreading platforms. Our studies concretely point towards the potential of our approach for the development of a large-scale lipreading MOOCs platform that can impact millions of people with hearing loss.
9) Lip-to-Speech Synthesis for Arbitrary Speakers in the Wild (ACM Multimedia 2022)
In this work, we address the problem of generating speech from silent lip videos for any speaker in the wild. In stark contrast to previous works, our method (i) is not restricted to a fixed number of speakers, (ii) does not explicitly impose constraints on the domain or the vocabulary and (iii) deals with videos that are recorded in the wild as opposed to within laboratory settings. The task presents a host of challenges, with the key one being that many features of the desired target speech, like voice, pitch and linguistic content, cannot be entirely inferred from the silent face video. In order to handle these stochastic variations, we propose a new VAE-GAN architecture that learns to associate the lip and speech sequences amidst the variations. With the help of multiple powerful discriminators that guide the training process, our generator learns to synthesize speech sequences in any voice for the lip movements of any person. Extensive experiments on multiple datasets show that we outperform all baselines by a large margin. Further, our network can be fine-tuned on videos of specific identities to achieve a performance comparable to single-speaker models that are trained on 4× more data. We conduct numerous ablation studies to analyze the effect of different modules of our architecture. We also provide a demo video that demonstrates several qualitative results along with the code and trained models on our website
10) Towards Automatic Speech to Sign Language Generation (Interspeech 2021)
We aim to solve the highly challenging task of generating continuous sign language videos solely from speech segments for the first time. Recent efforts in this space have focused on generating such videos from human-annotated text transcripts without considering other modalities. However, replacing speech with sign language proves to be a practical solution while communicating with people suffering from hearing loss. Therefore, we eliminate the need of using text as input and design techniques that work for more natural, continuous, freely uttered speech covering an extensive vocabulary. Since the current datasets are inadequate for generating sign language directly from speech, we collect and release the first Indian sign language dataset comprising speech-level annotations, text transcripts, and the corresponding sign-language videos. Next, we propose a multi-tasking transformer network trained to generate signer’s poses from speech segments. With speech-to-text as an auxiliary task and an additional cross-modal discriminator, our model learns to generate continuous sign pose sequences in an end-to-end manner. Extensive experiments and comparisons with other baselines demonstrate the effectiveness of our approach. We also conduct additional ablation studies to analyze the effect of different modules of our network. A demo video containing several results is attached to the supplementary material.
11) Translating Sign Language to Talking Faces (ICVGIP 2021)
Communication with the deaf community relies profoundly on the interpretation of sign languages performed by the signers. In light of the recent breakthroughs in sign language translations, we propose a pipeline that we term “Translating Sign Language Videos to Talking Faces”. In this context, we improve the existing sign language translation systems by using POS tags to improve language modeling. We further extend the challenge to develop a system that can interpret a video from a signer to an avatar speaking in spoken languages. We focus on the translation systems that attempt to translate sign languages to text without glosses, an expensive annotation form. We critically analyze two state-of-the-art architectures, and based on their limitations, we improvise the systems. We propose a two-stage approach to translate sign language into intermediate text followed by a language model to get the final predictions. Quantitative evaluations on the challenging benchmarks on RWTH-PHOENIX-Weather 2014 T show that the translation accuracy of the texts generated by our translation model improves the state-of-the-art models by approximately 3 points. We then build a working text to talking face generation pipeline by bringing together multiple existing modules. The overall pipeline is capable of generating talking face videos with speech from sign language poses.
12) Intelligent Video Editor for Talking Faces (ICVGIP 2021)
This paper proposes a video editor based on OpenShot with several state-of-the-art facial video editing algorithms as added functionalities. Our editor provides an easy-to-use interface to apply modern lip-syncing algorithms interactively. Apart from lip-syncing, the editor also uses audio and facial re-enactment to generate expressive talking faces. The manual control improves the overall experience of video editing without missing out on the benefits of modern synthetic video generation algorithms. This control enables us to lip-sync complex dubbed movie scenes, interviews, television shows, and other visual content. Furthermore, our editor provides features that automatically translate lectures from spoken content, lip-sync of the professor, and background content like slides. While doing so, we also tackle the critical aspect of synchronizing background content with the translated speech. We qualitatively evaluate the usefulness of the proposed editor by conducting human evaluations. Our evaluations show a clear improvement in the efficiency of using human editors and an improved video generation quality.
13) Audio-Visual Face Reenactment (WACV 2023)
This work proposes a novel method to generate realistic talking head videos using audio and visual streams. We animate a source image by transferring head motion from a driving video using a dense motion field generated using learnable keypoints. We improve the quality of lip sync using audio as an additional input, helping the network to attend to the mouth region. We use additional priors using face segmentation and face mesh to improve the structure of the reconstructed faces. Finally, we improve the visual quality of the generations by incorporating a carefully designed identity-aware generator module. The identity-aware generator takes the source image and the warped motion features as input to generate a high-quality output with fine-grained details. Our method produces state-of-the-art results and generalizes well to unseen faces, languages, and voices. We comprehensively evaluate our approach using multiple metrics and outperforming the current techniques both qualitative and quantitatively. Our work opens up several applications, including enabling low bandwidth video calls. A demo video is attached to the supplementary material demonstrating our approach’s results and effectiveness.
14) Towards Generating Ultra-High Resolution Talking-Face Videos with Lip Synchronisation (WACV 2023)
Talking-face video generation works have achieved state-of-the-art results in synthesizing videos with lip synchronization. However, most of the previous works deal with low-resolution talking-face videos (up to 256x256 pixels), thus, generating extremely high-resolution videos still remains a challenge. We take a giant leap in this work and propose a novel method to synthesize talking-face videos at resolutions as high as 4K! Our task presents several key challenges: (i) Scaling the existing methods to such high resolutions is resource-constrained, both in terms of compute and the availability of very high-resolution datasets, (ii) The synthesized videos need to be spatially and temporally coherent. The sheer number of pixels that the model needs to generate while maintaining the temporal consistency at the video level makes this task non-trivial and has never been attempted before in literature. To address these issues, we propose to train the lip-sync generator in a compact Vector Quantized (VQ) space for the first time. Our core idea to encode the faces in a compact 16x16 representation allows us to model high-resolution videos. In our framework, we learn the lip movements in the quantized space on the newly collected 4K Talking Faces (4KTF) dataset. Our approach is speaker agnostic and can handle various languages and voices. We benchmark our technique against several competitive works and show that we can achieve a remarkable 64-times more pixels than the current state-of-the-art! We also ensure our model implicitly watermarks the generated videos to mitigate any potential misuse. Our supplementary demo video depicts additional qualitative results, comparisons, and several real-world applications, like professional movie editing enabled by our model.
15) Audio-Visual Zero Shot Learning (WACV 2021)
In this work, we propose a novel approach for generalized zero-shot learning in a multi-modal setting, where we have novel classes of audio/video during testing that are not seen during training. We use the semantic relatedness of text embeddings as a means for zero-shot learning by aligning audio and video embeddings with the corresponding class label text feature space. Our approach uses a cross-modal decoder and a composite triplet loss. The cross-modal decoder enforces a constraint that the class label text features can be reconstructed from the audio and video embeddings of data points. This helps the audio and video embeddings to move closer to the class label text embedding. The composite triplet loss makes use of the audio, video, and text embeddings. It helps bring the embeddings from the same class closer and push away the embeddings from different classes in a multi-modal setting. This helps the network to perform better on the multi-modal zero-shot learning task. Importantly, our multi-modal zero-shot learning approach works even if a modality is missing at test time. We test our approach on the generalized zero-shot classification and retrieval tasks and show that our approach outperforms other models in the presence of a single modality as well as in the presence of multiple modalities. We validate our approach by comparing it with previous approaches and using various ablations.
16) Audio-Visual Generation (ICASSP 2021)
There have been a number of techniques that have demonstrated the generation of multimedia data for one modality at a time using GANs, such as the ability to generate images, videos, and audio. However, so far, the task of multi-modal generation of data, specifically for audio and videos both, has not been sufficiently well-explored. Towards this, we propose a method that demonstrates that we are able to generate naturalistic samples of video and audio data by the joint correlated generation of audio and video modalities. The proposed method uses multiple discriminators to ensure that the audio, video, and the joint output are also indistinguishable from real-world samples. We present a dataset for this task and show that we are able to generate realistic samples. This method is validated using various standard metrics such as Inception Score, Frechet Inception Distance (FID) and through human evaluation.
17) Stochastic Audio to Video Generation (Interspeech 2020)
The ability to envisage the visual of a talking face based just on hearing a voice is a unique human capability. There have been a number of works that have solved for this ability recently. We differ from these approaches by enabling a variety of talking face generations based on single audio input. Indeed, just having the ability to generate a single talking face would make a system almost robotic in nature. In contrast, our unsupervised stochastic audio-to-video generation model allows for diverse generations from a single audio input. Particularly, we present an unsupervised stochastic audio-to-video generation model that can capture multiple modes of the video distribution. We ensure that all the diverse generations are plausible. We do so through a principled multi-modal variational autoencoder framework. We demonstrate its efficacy on the challenging LRW and GRID datasets and demonstrate performance better than the baseline, while having the ability to generate multiple diverse lip synchronized videos.
18) Speech Prediction in Silent Videos using Variational Autoencoders (ICASSP 2021)
Understanding the relationship between the auditory and visual signals is crucial for many different applications ranging from computer-generated imagery (CGI) and video editing automation to assisting people with hearing or visual impairments. However, this is challenging since the distribution of both audio and visual modality is inherently multimodal. Therefore, most of the existing methods ignore the multimodal aspect and assume that there only exists a deterministic one-to-one mapping between the two modalities. It can lead to low-quality predictions as the model collapses to optimizing the average behavior rather than learning the full data distributions. In this paper, we present a stochastic model for generating speech in a silent video. The proposed model combines recurrent neural networks and variational deep generative models to learn the auditory signal’s conditional distribution given the visual signal. We demonstrate the performance of our model on the GRID dataset based on standard benchmarks.
Future Work
We are presently exploring inclusion of 3D information for the Audio-Visual Humans project.