Models for low-latency, streaming applications could benefit from the knowledge capacity of larger models, but edge devices cannot run these models due to resource constraints. A possible solution is to transfer hints during inference from a large model running remotely to a small model running…
Accurate control of the total duration of generated speech by adjusting the speech rate is crucial for various text-to-speech (TTS) applications. However, the impact of adjusting the speech rate on speech quality, such as intelligibility and speaker characteristics, has been underexplored. In this work, we…
Given the popularity of generative AI, Large Language Models (LLMs) often consume hundreds or thousands of GPUs for parallelizing and accelerating the training process. Communication overhead becomes more pronounced when training LLMs at scale. To eliminate communication overhead in distributed LLM training, we propose Domino,…
Recently, zero-shot text-to-speech (TTS) systems, capable of synthesizing any speaker's voice from a short audio prompt, have made rapid advancements. However, the quality of the generated speech significantly deteriorates when the audio prompt contains noise, and limited research has been conducted to address this issue.…
Abstract to come...
Among the many narratives of the transformative power of Generative AI is one that sees in the world a latent nation of programmers who need to wield nothing but intentions and natural language to render their ideas in software. In this paper, this outlook is…
Public speaking experts intentionally take their audience on an emotional roller coaster, staying attuned to their audience's collective emotional feedback. In this research, we explore how bidirectional sharing of heart rates between a speaker and their audience facilitates this emotional exchange, through empathy, emotional awareness,…
Background/objectives Pancreatic cyst management can be distilled into three separate pathways – discharge, monitoring or surgery– based on the risk of malignant transformation. This study compares the performance of artificial intelligence (AI) models to clinical care for this task. Methods Two explainable boosting machine (EBM)…
While there is excitement about the potential of algorithms to optimize individual decision-making, changing individual behavior will, almost inevitably, impact markets. Yet little is known about these effects. In this paper, I study how the availability of algorithmic prediction changes entry, allocation, and prices in…
Large language model (LLM)-based AI delegates are increasingly utilized to act on behalf of users, assisting them with a wide range of tasks through conversational interfaces. Despite their advantages, concerns arise regarding the potential risk of privacy leaks, particularly in scenarios involving social interactions. While…
Understanding node representations in graph-based models is crucial for uncovering biases, diagnosing errors, and building trust in model decisions. However, previous work on explainable AI for node representations has primarily emphasized explanations (reasons for model predictions) rather than interpretations (mapping representations to understandable concepts). Furthermore,…
While audio quality is a key performance metric for various audio processing tasks, including generative modeling, its objective measurement remains a challenge. Audio-Language Models (ALMs) are pre-trained on audio-text pairs that may contain information about audio quality, the presence of artifacts, or noise. Given an…
