We introduce a large language model (LLM) capable of processing speech inputs and show that tuning it further with reinforcement learning on human preference (RLHF) enables it to adapt better to disordered speech than traditional fine-tuning. Our method replaces low-frequency text tokens in an LLM’s vocabulary with audio tokens and enables the model to recognize speech by fine-tuning it on speech with transcripts. We then use RL with rewards based on syntactic and semantic accuracy measures generalizing the LLM further to recognize disordered speech. While the resulting LLM does not outperform existing systems for speech recognition we find that tuning with reinforcement learning using custom rewards leads to substantially better performance than supervised fine-tuning of the language model, specifically when adapting to speech in a different setting. This presents a compelling alternative tuning strategy for speech recognition using large language models.

This study investigates the impact of integrating a dataset of disordered speech recordings (~1,000 hours) into the fine-tuning of a near state-of-the-art ASR baseline system. Contrary to what one might expect, despite the data being less than 1% of the training data of the ASR system, we find a considerable improvement in disordered speech recognition accuracy. Specifically, we observe a 33% improvement on prompted speech, and a 26% improvement on a newly gathered spontaneous, conversational dataset of disordered speech. Importantly, there is no significant performance decline on standard speech recognition benchmarks. Further, we observe that the proposed tuning strategy helps close the gap between the baseline system and personalized models by 64% highlighting the significant progress as well as the room for improvement. Given the substantial benefits of our findings, this experiment suggests that from a fairness perspective, incorporating a small fraction of high quality disordered speech data in a training recipe is an easy step that could be done to make speech technology more accessible for users with speech disabilities.

In our era of rapid technological advancement, the research landscape for writing assistants has become increasingly fragmented across various research communities. We seek to address this challenge by proposing a design space as a structured way to examine and explore the multidimensional space of intelligent and interactive writing assistants. Through community collaboration, we explore five aspects of writing assistants: task, user, technology, interaction, and ecosystem. Within each aspect, we define dimensions and codes by systematically reviewing 115 papers, while leveraging the expertise of researchers in various disciplines. Our design space aims to offer researchers and designers a practical tool to navigate, comprehend, and compare the various possibilities of writing assistants, and aid in the design of new writing assistants.

Automatic Speech Recognition (ASR) systems, despite significant advances in recent years, still have much room for improvement particularly in the recognition of disordered speech. Even so, erroneous transcripts from ASR models can help people with disordered speech be better understood, especially if the transcription doesn’t significantly change the intended meaning. Evaluating the efficacy of ASR for this use case requires a methodology for measuring the impact of transcription errors on the intended meaning and comprehensibility. Human evaluation is the gold standard for this, but it can be laborious, slow, and expensive. In this work, we tune and evaluate large language models for this task and find them to be a much better proxy for human evaluators than other metrics commonly used. We further present a case-study using the presented approach to assess the quality of personalized ASR models to make model deployment decisions and correctly set user expectations for model quality as part of our trusted tester program.

Seeking answers to questions within long scientific research articles is a crucial area of study that aids readers in quickly addressing their inquiries. However, existing question-answering (QA) datasets based on scientific papers are limited in scale and focus solely on textual content. We introduce SPIQA (Scientific Paper Image Question Answering), the first large-scale QA dataset specifically designed to interpret complex figures and tables within the context of scientific research articles across various domains of computer science. Leveraging the breadth of expertise and ability of multimodal large language models (MLLMs) to understand figures, we employ automatic and manual curation to create the dataset. We craft an information-seeking task on interleaved images and text that involves multiple images covering a wide variety of plots, charts, tables, schematic diagrams, and result visualizations. SPIQA comprises 270K questions divided into training, validation, and three different evaluation splits. Through extensive experiments with 12 prominent foundational models, we evaluate the ability of current multimodal systems to comprehend the nuanced aspects of research articles. Additionally, we propose a Chain-of-Thought (CoT) evaluation strategy with in-context retrieval that allows fine-grained, step-by-step assessment and improves model performance. We further explore the upper bounds of performance enhancement with additional textual information, highlighting its promising potential for future research and the dataset’s impact on revolutionizing how we interact with scientific literature.

Large Language Models (LLMs) may offer transformative opportunities for text input, especially for physically demanding modalities like handwriting. We studied a form of abbreviated handwriting by designing, developing, and evaluating a prototype, named SkipWriter, that converts handwritten strokes of a variable-length prefix-based abbreviation (e.g., "ho a y" as handwritten strokes) into the intended full phrase (e.g., "how are you" in the digital format) based on the preceding context. SkipWriter consists of an in-production handwriting recognizer and an LLM fine-tuned on this task. With flexible pen input, SkipWriter allows the user to add and revise prefix strokes when predictions do not match the user’s intent. An user evaluation demonstrated a 60% reduction in motor movements with an average speed of 25.78 WPM. We also showed that this reduction is close to the ceiling of our model in an offline simulation.

We developed dysarthric speech intelligibility classifiers on 551,176 disordered speech samples contributed by a diverse set of 468 speakers, with a range of self-reported speaking disorders and rated for their overall intelligibility on a five-point scale. We trained three models following different deep learning approaches and evaluated them on ~ 94K utterances from 100 speakers. We further found the models to generalize well (without further training) on the TORGO database[1] (100% accuracy), UASpeech[2] (0.93 correlation), ALS-TDI PMP[3] (0.81 AUC) datasets as well as on a dataset of realistic unprompted speech we gathered (106 dysarthric and 76 control speakers, ~ 2300 samples). We share our model1to advance research in this domain.

Recent advances in self-supervision have dramatically improved the quality of speech representations. However, deployment of state-of-the-art embedding models on devices has been restricted due to their limited public availability and large resource footprint. Our work addresses these issues by publicly releasing a collection of paralinguistic speech models that are small and near state-of-the-art performance. Our approach is based on knowledge distillation, and our models are distilled on public data only. We explore different architectures and thoroughly evaluate our models on the Non-Semantic Speech (NOSS) benchmark. Our largest distilled model is less than 15% the size of the original model (314MB vs 2.2GB), achieves over 96% the accuracy on 6 of 7 tasks, and is trained on 6.5% the data. The smallest model is 1% in size (22MB) and achieves over 90% the accuracy on 6 of 7 tasks. Our models outperform the open source Wav2Vec 2.0 model on 6 of 7 tasks, and our smallest model outperforms the open source Wav2Vec 2.0 on both emotion recognition tasks despite being 7% the size.

Shanqing Cai, Subhashini Venugopalan, Katrin Tomanek, Ajit Narayanan, Meredith Morris, Michael Brenner. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Improving the performance of deep networks in data-limited regimes has warranted much attention. In this work, we empirically show that “winning tickets” (small sub-networks) obtained via magnitude pruning based on the lottery ticket hypothesis, apart from being sparse are also effective recognizers in data-limited regimes. Based on extensive experiments, we find that in low data regimes (datasets of 50-100 examples per class), sparse winning tickets substantially outperform the original dense networks. This approach, when combined with augmentations or fine-tuning from a self-supervised backbone network, shows further improvements in performance by as much as 16% (absolute) on low-sample datasets and long-tailed classification. Further, sparse winning tickets are more robust to synthetic noise and distribution shifts compared to their dense counterparts. Our analysis of winning tickets on small datasets indicates that, though sparse, the networks retain density in the initial layers and their representations are more generalizable. Code is available at https://github.com/VITA-Group/DataEfficientLTH.

Integrated Gradients (IG) [29] is a commonly used feature attribution method for deep neural networks. While IG has many desirable properties, the method often produces spurious/noisy pixel attributions in regions that are not related to the predicted class when applied to visual models. While this has been previously noted [27], most existing solutions [25], [17] are aimed at addressing the symptoms by explicitly reducing the noise in the resulting attributions. In this work, we show that one of the causes of the problem is the accumulation of noise along the IG path. To minimize the effect of this source of noise, we propose adapting the attribution path itself - conditioning the path not just on the image but also on the model being explained. We introduce Adaptive Path Methods (APMs) as a generalization of path methods, and Guided IG as a specific instance of an APM. Empirically, Guided IG creates saliency maps better aligned with the model's prediction and the input image that is being explained. We show through qualitative and quantitative experiments that Guided IG outperforms other, related methods in nearly every experiment.

Automatic classification of disordered speech can provide an objective tool for identifying the presence and severity of speech impairment. Classification approaches can also help identify hard-to-recognize speech samples to teach ASR systems about the variable manifestations of impaired speech. Here, we develop and compare different deep learning techniques to classify the intelligibility of disordered speech on selected phrases. We collected samples from a diverse set of 661 speakers with a variety of self-reported disorders speaking 29 words or phrases, which were rated by speech-language pathologists for their overall intelligibility using a five-point Likert scale. We then evaluated classifiers developed using 3 approaches: (1) a convolutional neural network (CNN) trained for the task, (2) classifiers trained on non-semantic speech representations from CNNs that used an unsupervised objective [1], and (3) classifiers trained on the acoustic (encoder) embeddings from an ASR system trained on typical speech [2]. We found that the ASR encoder's embeddings considerably outperform the other two on detecting and classifying disordered speech. Further analysis shows that the ASR embeddings cluster speech by the spoken phrase, while the non-semantic embeddings cluster speech by speaker. Also, longer phrases are more indicative of intelligibility deficits than single words.

We study the attribution problem for deep networks applied to perception tasks. For vision tasks, attribution techniques attribute the prediction of a network to the pixels of the input image. We propose a new technique called Blur Integrated Gradients (Blur IG). This technique has several advantages over other methods. First, it can tell at what scale a network recognizes an object. It produces scores in the scale/frequency dimension, that we find captures interesting phenomena. Second, it satisfies the scale-space axioms, which imply that it employs perturbations that are free of artifact. We therefore produce explanations that are cleaner and consistent with the operation of deep networks. Third, it eliminates the need for baseline parameter for Integrated Gradients for perception tasks. This is desirable because the choice of baseline has a significant effect on the explanations. We compare the proposed technique against previous techniques and demonstrate application on three tasks: ImageNet object recognition, Diabetic Retinopathy prediction, and AudioSet audio event identification. Code and examples are at https://github.com/PAIR-code/saliency.

MOTIVATION: Advances in automation and imaging have made it possible to capture a large image dataset that spans multiple experimental batches of data. However, accurate biological comparison across the batches is challenged by batch-to-batch variation (i.e. batch effect) due to uncontrollable experimental noise (e.g. varying stain intensity or cell density). Previous approaches to minimize the batch effect have commonly focused on normalizing the low-dimensional image measurements such as an embedding generated by a neural network. However, normalization of the embedding could suffer from over-correction and alter true biological features (e.g. cell size) due to our limited ability to interpret the effect of the normalization on the embedding space. Although techniques like flat-field correction can be applied to normalize the image values directly, they are limited transformations that handle only simple artifacts due to batch effect. RESULTS: We present a neural network-based batch equalization method that can transfer images from one batch to another while preserving the biological phenotype. The equalization method is trained as a generative adversarial network (GAN), using the StarGAN architecture that has shown considerable ability in style transfer. After incorporating new objectives that disentangle batch effect from biological features, we show that the equalized images have less batch information and preserve the biological information. We also demonstrate that the same model training parameters can generalize to two dramatically different types of cells, indicating this approach could be broadly applicable. AVAILABILITY AND IMPLEMENTATION: https://github.com/tensorflow/gan/tree/master/tensorflow_gan/examples/stargan. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Recent captioning models are limited in their ability to scale and describe concepts unseen in paired image-text corpora. We propose the Novel Object Captioner (NOC), a deep visual semantic captioning model that can describe a large number of object categories not present in existing image-caption datasets. Our model takes advantage of external sources - labeled images from object recognition datasets, and semantic knowledge extracted from unannotated text. We propose minimizing a joint objective which can learn from these diverse data sources and leverage distributional semantic embeddings, enabling the model to generalize and describe novel objects outside of image-caption datasets. We demonstrate that our model exploits semantic information to generate captions for hundreds of object categories in the ImageNet object recognition dataset that are not observed in MSCOCO image-caption training data, as well as many categories that are observed very rarely. Both automatic evaluations and human judgements show that our model considerably outperforms prior work in being able to describe many more categories of objects.

Long videos captured by consumers are typically tied to some of the most important moments of their lives, yet ironically are often the least frequently watched. The time required to initially retrieve and watch sections can be daunting. In this work we propose novel techniques for summarizing and annotating long videos. Existing video summarization techniques focus exclusively on identifying keyframes and subshots, however evaluating these summarized videos is a challenging task. Our work proposes methods to generate visual summaries of long videos, and in addition proposes techniques to annotate and generate textual summaries of the videos using recurrent networks. Interesting segments of long video are extracted based on image quality as well as cinematographic and consumer preference. Key frames from the most impactful segments are converted to textual annotations using sequential encoding and decoding deep learning models. Our summarization technique is benchmarked on the VideoSet dataset, and evaluated by humans for informative and linguistic content. We believe this to be the first fully automatic method capable of simultaneous visual and textual summarization of long consumer videos.

Models based on deep convolutional networks have dominated recent image interpretation tasks; we investigate whether models which are also recurrent are effective for tasks involving sequences, visual and otherwise. We describe a class of recurrent convolutional architectures which is end-to-end trainable and suitable for large-scale visual understanding tasks, and demonstrate the value of these models for activity recognition, image captioning, and video description. In contrast to previous models which assume a fixed visual representation or perform simple temporal averaging for sequential processing, recurrent convolutional models are "doubly deep" in that they learn compositional representations in space and time. Learning long-term dependencies is possible when nonlinearities are incorporated into the network state updates. Differentiable recurrent models are appealing in that they can directly map variable-length inputs (e.g., videos) to variable-length outputs (e.g., natural language text) and can model complex temporal dynamics; yet they can be optimized with backpropagation. Our recurrent sequence models are directly connected to modern visual convolutional network models and can be jointly trained to learn temporal dynamics and convolutional perceptual representations. Our results show that such models have distinct advantages over state-of-the-art models for recognition or generation which are separately defined or optimized.

While recent deep neural network models have achieved promising results on the image captioning task, they rely largely on the availability of corpora with paired image and sentence captions to describe objects in context. In this work, we propose the Deep Compositional Captioner (DCC) to address the task of generating descriptions of novel objects which are not present in paired imagesentence datasets. Our method achieves this by leveraging large object recognition datasets and external text corpora and by transferring knowledge between semantically similar concepts. Current deep caption models can only describe objects contained in paired image-sentence corpora, despite the fact that they are pre-trained with large object recognition datasets, namely ImageNet. In contrast, our model can compose sentences that describe novel objects and their interactions with other objects. We demonstrate our model's ability to describe novel concepts by empirically evaluating its performance on MSCOCO and show qualitative results on ImageNet images of objects for which no paired image-sentence data exist. Further, we extend our approach to generate descriptions of objects in video clips. Our results show that DCC has distinct advantages over existing image and video captioning approaches for generating descriptions of new objects in context.

This paper investigates how linguistic knowledge mined from large text corpora can aid the generation of natural language descriptions of videos.Specifically, we integrate both a neural language model and distributional semantics trained on large text corpora into a recent LSTM-based architecture for video description.We evaluate our approach on a collection of Youtube videos as well as two large movie description datasets showing significant improvements in grammaticality while modestly improving descriptive quality.

Real-world web videos often contain cues to supplement visual information for generating natural language descriptions. In this paper we propose a sequence-to-sequence model which explores such auxiliary information. In particular, audio and the topic of the video are used in addition to the visual information in a multimodal framework to generate coherent descriptions of videos "in the wild". In contrast to current encoder-decoder based models which exploit visual information only during the encoding stage, our model fuses multiple sources of information judiciously, showing improvement over using the different modalities separately. We based our multimodal video description network on the state-of-the-art sequence to sequence video to text (S2VT) model and extended it to take advantage of multiple modalities. Extensive experiments on the challenging MSR-VTT dataset are carried out to show the superior performance of the proposed approach on natural videos found in the web.

Models based on deep convolutional networks have dominated recent image interpretation tasks; we investigate whether models which are also recurrent, or “temporally deep”, are effective for tasks involving sequences, visual and otherwise. We develop a novel recurrent convolutional architecture suitable for large-scale visual learning which is end-to-end trainable, and demonstrate the value of these models on benchmark video recognition tasks, image description and retrieval problems, and video narration challenges. In contrast to current models which assume a fixed spatio-temporal receptive field or simple temporal averaging for sequential processing, recurrent convolutional models are “doubly deep” in that they can be compositional in spatial and temporal “layers”. Such models may have advantages when target concepts are complex and/or training data are limited. Learning long-term dependencies is possible when nonlinearities are incorporated into the network state updates. Long-term RNN models are appealing in that they directly can map variable-length inputs (e.g., video frames) to variable length outputs (e.g., natural language text) and can model complex temporal dynamics; yet they can be optimized with backpropagation. Our recurrent long-term models are directly connected to modern visual convnet models and can be jointly trained to simultaneously learn temporal dynamics and convolutional perceptual representations. Our results show such models have distinct advantages over state-of-the-art models for recognition or generation which are separately defined and/or optimized.

Real-world videos often have complex dynamics, methods for generating open-domain video descriptions should be sensitive to temporal structure and allow both input (sequence of frames) and output (sequence of words) of variable length. To approach this problem we propose a novel end-to-end sequence-to-sequence model to generate captions for videos. For this we exploit recurrent neural networks, specifically LSTMs, which have demonstrated state-of-the-art performance in image caption generation. Our LSTM model is trained on video-sentence pairs and learns to associate a sequence of video frames to a sequence of words in order to generate a description of the event in the video clip. Our model naturally is able to learn the temporal structure of the sequence of frames as well as the sequence model of the generated sentences, i.e. a language model. We evaluate several variants of our model that exploit different visual features on a standard set of YouTube videos and two movie description datasets (M-VAD and MPII-MD).

Subhashini Venugopalan, Huijuan Xu, Jeff Donahue, Marcus Rohrbach, Raymond Mooney, Kate Saenko. Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2015.

Despite a recent push towards large-scale object recognition, activity recognition remains limited to narrow domains and small vocabularies of actions. In this paper, we tackle the challenge of recognizing and describing activities ``in-the-wild''. We present a solution that takes a short video clip and outputs a brief sentence that sums up the main activity in the video, such as the actor, the action and its object. Unlike previous work, our approach works on out-of-domain actions: it does not require training videos of the exact activity. If it cannot find an accurate prediction for a pre-trained model, it finds a less specific answer that is also plausible from a pragmatic standpoint. We use semantic hierarchies learned from the data to help to choose an appropriate level of generalization, and priors learned from Web-scale natural language corpora to penalize unlikely combinations of actors/actions/objects, we also use a Web-scale language model to ``fill in'' novel verbs, i.e. when the verb does not appear in the training set. We evaluate our method on a large YouTube corpus and demonstrate it is able to generate short sentence descriptions of video clips better than baseline approaches.