Blog(4)

• Back Transcription as a Method for Evaluating Robustness of Natural Language Understanding Models to Speech Recognition Errors

  In a spoken dialogue system, an NLU model is preceded by a speech recognition system that can deteriorate the performance of natural language understanding. We propose to investigate the impact of speech recognition errors on the performance of natural language understanding models with a procedure that consists of three stages

• Multilingual Open Custom Keyword Spotting Testset

  In the ever-growing landscape of technology, voice intelligence-based solutions evolved and reshaped how we interact with devices, applications, and services.

• Pretrained Bidirectional Distillation for Machine Translation

  Initializing parameters by a pretrained masked language model (LM) [1] is a knowledge transfer method widely applied to natural language processing tasks. Following its success, pretrained neural machine translation (NMT) models have attracted more and more research interest [2,3,4,5].

View More

Research Areas(0)

Publications(29)

• HOP to the Next Tasks and Domains for Continual Learning in NLP

  AuthorUmberto Michieli, Mete Ozay

  PublishedNational Conference on Artificial Intelligence (AAAI)

  Date2024-03-24

• Joint Embedding Learning and Latent Subspace Probing for Cross-Domain Few-Shot Keyword Spotting

  AuthorMete Ozay

  PublishedInternational Conference on Acoustics, Speech and Signal Processing (ICASSP)

  Date2024-03-18

• BayesTune: Bayesian Sparse Deep Model Fine-tuning

  AuthorMinyoung Kim, Timothy Hospedales

  PublishedConference on Neural Information Processing Systems (NeurIPS)

  Date2023-12-11

View More

News(2)

• SRC-B Ranks Top Places in the Semantic Evaluation International Competition

  Natural language processing (NLP) aims to enhance computers to be capable of understanding the semantics of languages. With the development of deep learning, NLP has achieved great advances in recent years. In particular, NLP-related competitions have seen many new NLP algorithms and have become an unprecedented way to promote NLP research.

• Samsung Research at EMNLP 2018

  Research from Samsung Research will be presented in the oral and poster sessions of the 2018 Conference on Empirical Methods in Natural Language Processing (EMNLP 2018).

Others(0)