“Probabilistic Modeling”

Comparative Study of Transformer and RNN in Speech Applications

by
Comparative Study of Transformer and RNN in Speech Applications

Original link: https://arxiv.org/pdf/1909.06317.pdf Abstract Sequence-to-sequence models are widely used in end-to-end speech processing, such as Automatic Speech Recognition (ASR), Speech Translation (ST), and Text-to-Speech (TTS). This paper focuses on a novel sequence-to-sequence model called the Transformer, which has achieved state-of-the-art performance in neural machine translation and other natural language processing applications. We conducted an in-depth … Read more

Lag-Llama: Probabilistic Time Series Forecasting with Foundation Models

by
Lag-Llama: Probabilistic Time Series Forecasting with Foundation Models

Abstract arXiv:2310.08278v3 [cs.LG] February 8, 2024 Original paper link: https://arxiv.org/pdf/2310.08278 In recent years, foundation models have caused a paradigm shift in the field of machine learning due to their unprecedented zero-shot and few-shot generalization capabilities. However, despite the success of foundation models in fields such as natural language processing and computer vision, the development of … Read more