In this paper, we present a novel inference algorithm that improves the prediction of state-of-the-art ASR models using nearest-neighbor-based matching on an ...
This paper presents a novel method that allows adapting ASR systems (transducers and enc-dec) to dictionaries containing rare words and phrases without ...
Dec 6, 2023 · In this paper, we present a novel inference algorithm that improves the prediction of state-of-the-art ASR models using nearest-neighbor-based ...
Request PDF | On Jan 1, 2023, Ashish Mittal and others published Speech-enriched Memory for Inference-time Adaptation of ASR Models to Word Dictionaries ...
Oct 9, 2023 · Our paper, "Speech-enriched Memory for Inference-time Adaptation of ASR Models to Word Dictionaries" has been accepted for EMNLP 2023!
No information is available for this page. · Learn why
Speech-enriched Memory for Inference-time Adaptation of ASR Models to Word Dictionaries. Ashish Mittal; Sunita Sarawagi; et al. 2023; EMNLP 2023. Global RNN ...
Speech-enriched Memory for Inference-time Adaptation of ASR Models to Word Dictionaries · Improved Neural Network-based Multi-label Classification with Better ...
Oct 24, 2024 · Compared to text-to-text language generation models, speech introduces a new variable, namely pronunciation variability, which influences ASR.
This paper demonstrates that contextual adapters can be applied to any general purpose pretrained ASR model to improve personalization.