Self-supervised learning general-purpose audio representations have demonstrated high performance in a variety of tasks. Although they can be optimized for application by fine-tuning, even higher performance can be expected if they can be specialized to pre-train for an application. This paper explores the challenges and solutions in specializing general-purpose audio representations for a specific application using speech, a highly demanding field, as an example. We enhance Masked Modeling Duo (M2D), a general-purpose model, to close the performance gap with state-of-the-art (SOTA) speech models. To do so, we propose a new task, denoising distillation, to learn from fine-grained clustered features, and M2D for Speech (M2D-S), which jointly learns the denoising distillation task and M2D masked prediction task. Experimental results show that M2D-S performs comparably to or outperforms SOTA speech models on the SUPERB benchmark, demonstrating that M2D can specialize in a demanding field. Our code is available at: https://github.com/nttcslab/m2d/tree/master/speech
Expanding the language coverage of speech technology has the potential to improve access to information for many more people. However, current speech technology is restricted to about one hundred languages which is a small fraction of the over 7,000 languages spoken around the world. The Massively Multilingual Speech (MMS) project increases the number of supported languages by 10-40x, depending on the task. The main ingredients are a new dataset based on readings of publicly available religious texts and effectively leveraging self-supervised learning. We built pre-trained wav2vec 2.0 models covering 1,406 languages, a single multilingual automatic speech recognition model for 1,107 languages, speech synthesis models for the same number of languages, as well as a language identification model for 4,017 languages. Experiments show that our multilingual speech recognition model more than halves the word error rate of Whisper on 54 languages of the FLEURS benchmark while being trained on a small fraction of the labeled data.
In several regions across the globe, snow has a significant impact on hydrology. The amounts of water that infiltrate the ground and flow as runoff are driven by the melting of snow. Therefore, it is crucial to study the magnitude and effect of snowmelt. Snow droughts, resulting from reduced snow storage, can drastically impact the water supplies in basins where snow predominates, such as in the western United States. Hence, it is important to detect the time and severity of snow droughts efficiently. We propose Snow Drought Response Index or SnoDRI, a novel indicator that could be used to identify and quantify snow drought occurrences. Our index is calculated using cutting-edge ML algorithms from various snow-related variables. The self-supervised learning of an autoencoder is combined with mutual information in the model. In this study, we use random forests for feature extraction for SnoDRI and assess the importance of each variable. We use reanalysis data (NLDAS-2) from 1981 to 2021 for the Pacific United States to study the efficacy of the new snow drought index. We evaluate the index by confirming the coincidence of its interpretation and the actual snow drought incidents.
Self-supervised learning (SSL) models use only the intrinsic structure of a given signal, independent of its acoustic domain, to extract essential information from the input to an embedding space. This implies that the utility of such representations is not limited to modeling human speech alone. Building on this understanding, this paper explores the cross-transferability of SSL neural representations learned from human speech to analyze bio-acoustic signals. We conduct a caller discrimination analysis and a caller detection study on Marmoset vocalizations using eleven SSL models pre-trained with various pretext tasks. The results show that the embedding spaces carry meaningful caller information and can successfully distinguish the individual identities of Marmoset callers without fine-tuning. This demonstrates that representations pre-trained on human speech can be effectively applied to the bio-acoustics domain, providing valuable insights for future investigations in this field.
Recently, contrastive self-supervised learning, where the proximity of representations is determined based on the identities of samples, has made remarkable progress in unsupervised representation learning. SimSiam is a well-known example in this area, known for its simplicity yet powerful performance. However, it is known to be sensitive to changes in training configurations, such as hyperparameters and augmentation settings, due to its structural characteristics. To address this issue, we focus on the similarity between contrastive learning and the teacher-student framework in knowledge distillation. Inspired by the ensemble-based knowledge distillation approach, the proposed method, EnSiam, aims to improve the contrastive learning procedure using ensemble representations. This can provide stable pseudo labels, providing better performance. Experiments demonstrate that EnSiam outperforms previous state-of-the-art methods in most cases, including the experiments on ImageNet, which shows that EnSiam is capable of learning high-quality representations.
Research in self-supervised learning (SSL) with natural images has progressed rapidly in recent years and is now increasingly being applied to and benchmarked with datasets containing remotely sensed imagery. A common benchmark case is to evaluate SSL pre-trained model embeddings on datasets of remotely sensed imagery with small patch sizes, e.g., 32x32 pixels, whereas standard SSL pre-training takes place with larger patch sizes, e.g., 224x224. Furthermore, pre-training methods tend to use different image normalization preprocessing steps depending on the dataset. In this paper, we show, across seven satellite and aerial imagery datasets of varying resolution, that by simply following the preprocessing steps used in pre-training (precisely, image sizing and normalization methods), one can achieve significant performance improvements when evaluating the extracted features on downstream tasks -- an important detail overlooked in previous work in this space. We show that by following these steps, ImageNet pre-training remains a competitive baseline for satellite imagery based transfer learning tasks -- for example we find that these steps give +32.28 to overall accuracy on the So2Sat random split dataset and +11.16 on the EuroSAT dataset. Finally, we report comprehensive benchmark results with a variety of simple baseline methods for each of the seven datasets, forming an initial benchmark suite for remote sensing imagery.
Although supervised learning has been highly successful in improving the state-of-the-art in the domain of image-based computer vision in the past, the margin of improvement has diminished significantly in recent years, indicating that a plateau is in sight. Meanwhile, the use of self-supervised learning (SSL) for the purpose of natural language processing (NLP) has seen tremendous successes during the past couple of years, with this new learning paradigm yielding powerful language models. Inspired by the excellent results obtained in the field of NLP, self-supervised methods that rely on clustering, contrastive learning, distillation, and information-maximization, which all fall under the banner of discriminative SSL, have experienced a swift uptake in the area of computer vision. Shortly afterwards, generative SSL frameworks that are mostly based on masked image modeling, complemented and surpassed the results obtained with discriminative SSL. Consequently, within a span of three years, over $100$ unique general-purpose frameworks for generative and discriminative SSL, with a focus on imaging, were proposed. In this survey, we review a plethora of research efforts conducted on image-oriented SSL, providing a historic view and paying attention to best practices as well as useful software packages. While doing so, we discuss pretext tasks for image-based SSL, as well as techniques that are commonly used in image-based SSL. Lastly, to aid researchers who aim at contributing to image-focused SSL, we outline a number of promising research directions.
3D point clouds are a crucial type of data collected by LiDAR sensors and widely used in transportation applications due to its concise descriptions and accurate localization. Deep neural networks (DNNs) have achieved remarkable success in processing large amount of disordered and sparse 3D point clouds, especially in various computer vision tasks, such as pedestrian detection and vehicle recognition. Among all the learning paradigms, Self-Supervised Learning (SSL), an unsupervised training paradigm that mines effective information from the data itself, is considered as an essential solution to solve the time-consuming and labor-intensive data labelling problems via smart pre-training task design. This paper provides a comprehensive survey of recent advances on SSL for point clouds. We first present an innovative taxonomy, categorizing the existing SSL methods into four broad categories based on the pretexts' characteristics. Under each category, we then further categorize the methods into more fine-grained groups and summarize the strength and limitations of the representative methods. We also compare the performance of the notable SSL methods in literature on multiple downstream tasks on benchmark datasets both quantitatively and qualitatively. Finally, we propose a number of future research directions based on the identified limitations of existing SSL research on point clouds.
Text-to-speech(TTS) has undergone remarkable improvements in performance, particularly with the advent of Denoising Diffusion Probabilistic Models (DDPMs). However, the perceived quality of audio depends not solely on its content, pitch, rhythm, and energy, but also on the physical environment. In this work, we propose ViT-TTS, the first visual TTS model with scalable diffusion transformers. ViT-TTS complement the phoneme sequence with the visual information to generate high-perceived audio, opening up new avenues for practical applications of AR and VR to allow a more immersive and realistic audio experience. To mitigate the data scarcity in learning visual acoustic information, we 1) introduce a self-supervised learning framework to enhance both the visual-text encoder and denoiser decoder; 2) leverage the diffusion transformer scalable in terms of parameters and capacity to learn visual scene information. Experimental results demonstrate that ViT-TTS achieves new state-of-the-art results, outperforming cascaded systems and other baselines regardless of the visibility of the scene. With low-resource data (1h, 2h, 5h), ViT-TTS achieves comparative results with rich-resource baselines.~\footnote{Audio samples are available at \url{https://ViT-TTS.github.io/.}}
Social recommendation is gaining increasing attention in various online applications, including e-commerce and online streaming, where social information is leveraged to improve user-item interaction modeling. Recently, Self-Supervised Learning (SSL) has proven to be remarkably effective in addressing data sparsity through augmented learning tasks. Inspired by this, researchers have attempted to incorporate SSL into social recommendation by supplementing the primary supervised task with social-aware self-supervised signals. However, social information can be unavoidably noisy in characterizing user preferences due to the ubiquitous presence of interest-irrelevant social connections, such as colleagues or classmates who do not share many common interests. To address this challenge, we propose a novel social recommender called the Denoised Self-Augmented Learning paradigm (DSL). Our model not only preserves helpful social relations to enhance user-item interaction modeling but also enables personalized cross-view knowledge transfer through adaptive semantic alignment in embedding space. Our experimental results on various recommendation benchmarks confirm the superiority of our DSL over state-of-the-art methods. We release our model implementation at: https://github.com/HKUDS/DSL.
This paper proposes handling training data sparsity in speech-based automatic depression detection (SDD) using foundation models pre-trained with self-supervised learning (SSL). An analysis of SSL representations derived from different layers of pre-trained foundation models is first presented for SDD, which provides insight to suitable indicator for depression detection. Knowledge transfer is then performed from automatic speech recognition (ASR) and emotion recognition to SDD by fine-tuning the foundation models. Results show that the uses of oracle and ASR transcriptions yield similar SDD performance when the hidden representations of the ASR model is incorporated along with the ASR textual information. By integrating representations from multiple foundation models, state-of-the-art SDD results based on real ASR were achieved on the DAIC-WOZ dataset.
We present a new self-supervised paradigm on point cloud sequence understanding. Inspired by the discriminative and generative self-supervised methods, we design two tasks, namely point cloud sequence based Contrastive Prediction and Reconstruction (CPR), to collaboratively learn more comprehensive spatiotemporal representations. Specifically, dense point cloud segments are first input into an encoder to extract embeddings. All but the last ones are then aggregated by a context-aware autoregressor to make predictions for the last target segment. Towards the goal of modeling multi-granularity structures, local and global contrastive learning are performed between predictions and targets. To further improve the generalization of representations, the predictions are also utilized to reconstruct raw point cloud sequences by a decoder, where point cloud colorization is employed to discriminate against different frames. By combining classic contrast and reconstruction paradigms, it makes the learned representations with both global discrimination and local perception. We conduct experiments on four point cloud sequence benchmarks, and report the results on action recognition and gesture recognition under multiple experimental settings. The performances are comparable with supervised methods and show powerful transferability.
In this paper, we aim to optimize a contrastive loss with individualized temperatures in a principled and systematic manner for self-supervised learning. The common practice of using a global temperature parameter $\tau$ ignores the fact that ``not all semantics are created equal", meaning that different anchor data may have different numbers of samples with similar semantics, especially when data exhibits long-tails. First, we propose a new robust contrastive loss inspired by distributionally robust optimization (DRO), providing us an intuition about the effect of $\tau$ and a mechanism for automatic temperature individualization. Then, we propose an efficient stochastic algorithm for optimizing the robust contrastive loss with a provable convergence guarantee without using large mini-batch sizes. Theoretical and experimental results show that our algorithm automatically learns a suitable $\tau$ for each sample. Specifically, samples with frequent semantics use large temperatures to keep local semantic structures, while samples with rare semantics use small temperatures to induce more separable features. Our method not only outperforms prior strong baselines (e.g., SimCLR, CLIP) on unimodal and bimodal datasets with larger improvements on imbalanced data but also is less sensitive to hyper-parameters. To our best knowledge, this is the first methodical approach to optimizing a contrastive loss with individualized temperatures.
Molecular representation learning is a crucial task in predicting molecular properties. Molecules are often modeled as graphs where atoms and chemical bonds are represented as nodes and edges, respectively, and Graph Neural Networks (GNNs) have been commonly utilized to predict atom-related properties, such as reactivity and solubility. However, functional groups (subgraphs) are closely related to some chemical properties of molecules, such as efficacy, and metabolic properties, which cannot be solely determined by individual atoms. In this paper, we introduce a new model for molecular representation learning called the Atomic and Subgraph-aware Bilateral Aggregation (ASBA), which addresses the limitations of previous atom-wise and subgraph-wise models by incorporating both types of information. ASBA consists of two branches, one for atom-wise information and the other for subgraph-wise information. Considering existing atom-wise GNNs cannot properly extract invariant subgraph features, we propose a decomposition-polymerization GNN architecture for the subgraph-wise branch. Furthermore, we propose cooperative node-level and graph-level self-supervised learning strategies for ASBA to improve its generalization. Our method offers a more comprehensive way to learn representations for molecular property prediction and has broad potential in drug and material discovery applications. Extensive experiments have demonstrated the effectiveness of our method.
We improve low-resource ASR by integrating the ideas of multilingual training and self-supervised learning. Concretely, we leverage an International Phonetic Alphabet (IPA) multilingual model to create frame-level pseudo labels for unlabeled speech, and use these pseudo labels to guide hidden-unit BERT (HuBERT) based speech pretraining in a phonetically-informed manner. The experiments on the Multilingual Speech (MLS) Corpus show that the proposed approach consistently outperforms the standard HuBERT on all the target languages. Moreover, on 3 of the 4 languages, comparing to the standard HuBERT, the approach performs better, meanwhile is able to save supervised training data by 1.5k hours (75%) at most. Our approach outperforms most of the state of the arts, with much less pretraining data in terms of hours and language diversity. Compared to XLSR-53 and a retraining based multilingual method, our approach performs better with full and limited finetuning data scenarios.
Speech fluency/disfluency can be evaluated by analyzing a range of phonetic and prosodic features. Deep neural networks are commonly trained to map fluency-related features into the human scores. However, the effectiveness of deep learning-based models is constrained by the limited amount of labeled training samples. To address this, we introduce a self-supervised learning (SSL) approach that takes into account phonetic and prosody awareness for fluency scoring. Specifically, we first pre-train the model using a reconstruction loss function, by masking phones and their durations jointly on a large amount of unlabeled speech and text prompts. We then fine-tune the pre-trained model using human-annotated scoring data. Our experimental results, conducted on datasets such as Speechocean762 and our non-native datasets, show that our proposed method outperforms the baseline systems in terms of Pearson correlation coefficients (PCC). Moreover, we also conduct an ablation study to better understand the contribution of phonetic and prosody factors during the pre-training stage.
The emergence of social media as one of the main platforms for people to access news has enabled the wide dissemination of fake news. This has motivated numerous studies on automating fake news detection. Although there have been limited attempts at unsupervised fake news detection, their performance suffers due to not exploiting the knowledge from various modalities related to news records and due to the presence of various latent biases in the existing news datasets. To address these limitations, this work proposes an effective framework for unsupervised fake news detection, which first embeds the knowledge available in four modalities in news records and then proposes a novel noise-robust self-supervised learning technique to identify the veracity of news records from the multi-modal embeddings. Also, we propose a novel technique to construct news datasets minimizing the latent biases in existing news datasets. Following the proposed approach for dataset construction, we produce a Large-scale Unlabelled News Dataset consisting 419,351 news articles related to COVID-19, acronymed as LUND-COVID. We trained the proposed unsupervised framework using LUND-COVID to exploit the potential of large datasets, and evaluate it using a set of existing labelled datasets. Our results show that the proposed unsupervised framework largely outperforms existing unsupervised baselines for different tasks such as multi-modal fake news detection, fake news early detection and few-shot fake news detection, while yielding notable improvements for unseen domains during training.
Transformer-based speech self-supervised learning (SSL) models, such as HuBERT, show surprising performance in various speech processing tasks. However, huge number of parameters in speech SSL models necessitate the compression to a more compact model for wider usage in academia or small companies. In this study, we suggest to reuse attention maps across the Transformer layers, so as to remove key and query parameters while retaining the number of layers. Furthermore, we propose a novel masking distillation strategy to improve the student model's speech representation quality. We extend the distillation loss to utilize both masked and unmasked speech frames to fully leverage the teacher model's high-quality representation. Our universal compression strategy yields the student model that achieves phoneme error rate (PER) of 7.72% and word error rate (WER) of 9.96% on the SUPERB benchmark.
The recent emergence of Self-Supervised Learning (SSL) as a fundamental paradigm for learning image representations has, and continues to, demonstrate high empirical success in a variety of tasks. However, most SSL approaches fail to learn embeddings that capture hierarchical semantic concepts that are separable and interpretable. In this work, we aim to learn highly separable semantic hierarchical representations by stacking Joint Embedding Architectures (JEA) where higher-level JEAs are input with representations of lower-level JEA. This results in a representation space that exhibits distinct sub-categories of semantic concepts (e.g., model and colour of vehicles) in higher-level JEAs. We empirically show that representations from stacked JEA perform on a similar level as traditional JEA with comparative parameter counts and visualise the representation spaces to validate the semantic hierarchies.
There has been a growing interest in using deep learning models for processing long surgical videos, in order to automatically detect clinical/operational activities and extract metrics that can enable workflow efficiency tools and applications. However, training such models require vast amounts of labeled data which is costly and not scalable. Recently, self-supervised learning has been explored in computer vision community to reduce the burden of the annotation cost. Masked autoencoders (MAE) got the attention in self-supervised paradigm for Vision Transformers (ViTs) by predicting the randomly masked regions given the visible patches of an image or a video clip, and have shown superior performance on benchmark datasets. However, the application of MAE in surgical data remains unexplored. In this paper, we first investigate whether MAE can learn transferrable representations in surgical video domain. We propose SurgMAE, which is a novel architecture with a masking strategy based on sampling high spatio-temporal tokens for MAE. We provide an empirical study of SurgMAE on two large scale long surgical video datasets, and find that our method outperforms several baselines in low data regime. We conduct extensive ablation studies to show the efficacy of our approach and also demonstrate it's superior performance on UCF-101 to prove it's generalizability in non-surgical datasets as well.