Chronic pain is a pervasive disorder which is often very disabling and is associated with comorbidities such as depression and anxiety. Neuropathic Pain (NP) is a common sub-type which is often caused due to nerve damage and has a known pathophysiology. Another common sub-type is Fibromyalgia (FM) which is described as musculoskeletal, diffuse pain that is widespread through the body. The pathophysiology of FM is poorly understood, making it very hard to diagnose. Standard medications and treatments for FM and NP differ from one another and if misdiagnosed it can cause an increase in symptom severity. To overcome this difficulty, we propose a novel framework, PainPoints, which accurately detects the sub-type of pain and generates clinical notes via summarizing the patient interviews. Specifically, PainPoints makes use of large language models to perform sentence-level classification of the text obtained from interviews of FM and NP patients with a reliable AUC of 0.83. Using a sufficiency-based interpretability approach, we explain how the fine-tuned model accurately picks up on the nuances that patients use to describe their pain. Finally, we generate summaries of these interviews via expert interventions by introducing a novel facet-based approach. PainPoints thus enables practitioners to add/drop facets and generate a custom summary based on the notion of "facet-coverage" which is also introduced in this work.
End-to-end Speech Translation (ST) aims at translating the source language speech into target language text without generating the intermediate transcriptions. However, the training of end-to-end methods relies on parallel ST data, which are difficult and expensive to obtain. Fortunately, the supervised data for automatic speech recognition (ASR) and machine translation (MT) are usually more accessible, making zero-shot speech translation a potential direction. Existing zero-shot methods fail to align the two modalities of speech and text into a shared semantic space, resulting in much worse performance compared to the supervised ST methods. In order to enable zero-shot ST, we propose a novel Discrete Cross-Modal Alignment (DCMA) method that employs a shared discrete vocabulary space to accommodate and match both modalities of speech and text. Specifically, we introduce a vector quantization module to discretize the continuous representations of speech and text into a finite set of virtual tokens, and use ASR data to map corresponding speech and text to the same virtual token in a shared codebook. This way, source language speech can be embedded in the same semantic space as the source language text, which can be then transformed into target language text with an MT module. Experiments on multiple language pairs demonstrate that our zero-shot ST method significantly improves the SOTA, and even performers on par with the strong supervised ST baselines.
The rapid development of deep natural language processing (NLP) models for text classification has led to an urgent need for a unified understanding of these models proposed individually. Existing methods cannot meet the need for understanding different models in one framework due to the lack of a unified measure for explaining both low-level (e.g., words) and high-level (e.g., phrases) features. We have developed a visual analysis tool, DeepNLPVis, to enable a unified understanding of NLP models for text classification. The key idea is a mutual information-based measure, which provides quantitative explanations on how each layer of a model maintains the information of input words in a sample. We model the intra- and inter-word information at each layer measuring the importance of a word to the final prediction as well as the relationships between words, such as the formation of phrases. A multi-level visualization, which consists of a corpus-level, a sample-level, and a word-level visualization, supports the analysis from the overall training set to individual samples. Two case studies on classification tasks and comparison between models demonstrate that DeepNLPVis can help users effectively identify potential problems caused by samples and model architectures and then make informed improvements.
Deep learning-based methods have made impressive progress in enhancing extremely low-light images - the image quality of the reconstructed images has generally improved. However, we found out that most of these methods could not sufficiently recover the image details, for instance, the texts in the scene. In this paper, a novel image enhancement framework is proposed to precisely restore the scene texts, as well as the overall quality of the image simultaneously under extremely low-light images conditions. Mainly, we employed a self-regularised attention map, an edge map, and a novel text detection loss. In addition, leveraging synthetic low-light images is beneficial for image enhancement on the genuine ones in terms of text detection. The quantitative and qualitative experimental results have shown that the proposed model outperforms state-of-the-art methods in image restoration, text detection, and text spotting on See In the Dark and ICDAR15 datasets.
Non-parallel many-to-many voice conversion remains an interesting but challenging speech processing task. Recently, AutoVC, a conditional autoencoder based method, achieved excellent conversion results by disentangling the speaker identity and the speech content using information-constraining bottlenecks. However, due to the pure autoencoder training method, it is difficult to evaluate the separation effect of content and speaker identity. In this paper, a novel voice conversion framework, named $\boldsymbol T$ext $\boldsymbol G$uided $\boldsymbol A$utoVC(TGAVC), is proposed to more effectively separate content and timbre from speech, where an expected content embedding produced based on the text transcriptions is designed to guide the extraction of voice content. In addition, the adversarial training is applied to eliminate the speaker identity information in the estimated content embedding extracted from speech. Under the guidance of the expected content embedding and the adversarial training, the content encoder is trained to extract speaker-independent content embedding from speech. Experiments on AIShell-3 dataset show that the proposed model outperforms AutoVC in terms of naturalness and similarity of converted speech.
The cover is the face of a book and is a point of attraction for the readers. Designing book covers is an essential task in the publishing industry. One of the main challenges in creating a book cover is representing the theme of the book's content in a single image. In this research, we explore ways to produce a book cover using artificial intelligence based on the fact that there exists a relationship between the summary of the book and its cover. Our key motivation is the application of text-to-image synthesis methods to generate images from given text or captions. We explore several existing text-to-image conversion techniques for this purpose and propose an approach to exploit these frameworks for producing book covers from provided summaries. We construct a dataset of English books that contains a large number of samples of summaries of existing books and their cover images. In this paper, we describe our approach to collecting, organizing, and pre-processing the dataset to use it for training models. We apply different text-to-image synthesis techniques to generate book covers from the summary and exhibit the results in this paper.
Current language models are considered to have sub-human capabilities at natural language tasks like question-answering or writing code. However, language models are not trained to perform well at these tasks, they are trained to accurately predict the next token given previous tokes in tokenized text. It is not clear whether language models are better or worse than humans at next token prediction. To try to answer this question, we performed two distinct experiments to directly compare humans and language models on this front: one measuring top-1 accuracy and the other measuring perplexity. In both experiments, we find humans to be consistently \emph{worse} than even relatively small language models like GPT3-Ada at next-token prediction.
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.
Neural Radiance Fields (NeRF) has achieved impressive results in single object scene reconstruction and novel view synthesis, which have been demonstrated on many single modality and single object focused indoor scene datasets like DTU, BMVS, and NeRF Synthetic.However, the study of NeRF on large-scale outdoor scene reconstruction is still limited, as there is no unified outdoor scene dataset for large-scale NeRF evaluation due to expensive data acquisition and calibration costs. In this paper, we propose a large-scale outdoor multi-modal dataset, OMMO dataset, containing complex land objects and scenes with calibrated images, point clouds and prompt annotations. Meanwhile, a new benchmark for several outdoor NeRF-based tasks is established, such as novel view synthesis, surface reconstruction, and multi-modal NeRF. To create the dataset, we capture and collect a large number of real fly-view videos and select high-quality and high-resolution clips from them. Then we design a quality review module to refine images, remove low-quality frames and fail-to-calibrate scenes through a learning-based automatic evaluation plus manual review. Finally, a number of volunteers are employed to add the text descriptions for each scene and key-frame to meet the potential multi-modal requirements in the future. Compared with existing NeRF datasets, our dataset contains abundant real-world urban and natural scenes with various scales, camera trajectories, and lighting conditions. Experiments show that our dataset can benchmark most state-of-the-art NeRF methods on different tasks. We will release the dataset and model weights very soon.
End-to-end (E2E) models have become the default choice for state-of-the-art speech recognition systems. Such models are trained on large amounts of labelled data, which are often not available for low-resource languages. Techniques such as self-supervised learning and transfer learning hold promise, but have not yet been effective in training accurate models. On the other hand, collecting labelled datasets on a diverse set of domains and speakers is very expensive. In this work, we demonstrate an inexpensive and effective alternative to these approaches by ``mining'' text and audio pairs for Indian languages from public sources, specifically from the public archives of All India Radio. As a key component, we adapt the Needleman-Wunsch algorithm to align sentences with corresponding audio segments given a long audio and a PDF of its transcript, while being robust to errors due to OCR, extraneous text, and non-transcribed speech. We thus create Shrutilipi, a dataset which contains over 6,400 hours of labelled audio across 12 Indian languages totalling to 4.95M sentences. On average, Shrutilipi results in a 2.3x increase over publicly available labelled data. We establish the quality of Shrutilipi with 21 human evaluators across the 12 languages. We also establish the diversity of Shrutilipi in terms of represented regions, speakers, and mentioned named entities. Significantly, we show that adding Shrutilipi to the training set of Wav2Vec models leads to an average decrease in WER of 5.8\% for 7 languages on the IndicSUPERB benchmark. For Hindi, which has the most benchmarks (7), the average WER falls from 18.8% to 13.5%. This improvement extends to efficient models: We show a 2.3% drop in WER for a Conformer model (10x smaller than Wav2Vec). Finally, we demonstrate the diversity of Shrutilipi by showing that the model trained with it is more robust to noisy input.