2021 IEEE Automatic Speech Recognition and Understanding Workshop (ASRU) 2021
DOI: 10.1109/asru51503.2021.9688247
|View full text |Cite
|
Sign up to set email alerts
|

Word-Level Confidence Estimation for RNN Transducers

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1

Citation Types

0
2
0

Year Published

2022
2022
2024
2024

Publication Types

Select...
3
2

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(2 citation statements)
references
References 15 publications
0
2
0
Order By: Relevance
“…We provided this list in the prompt and instructed the model to pick the most relevant phrase, i.e., "Recognize this speech in language English using potential mention -{biasing entity}". We used an off-the-shelf speech retriever [5] to retrieve top-1 entity mention from the speech, and replace {biasing entity} with it in the prompt above.…”
Section: Speech Translationmentioning
confidence: 99%
See 1 more Smart Citation
“…We provided this list in the prompt and instructed the model to pick the most relevant phrase, i.e., "Recognize this speech in language English using potential mention -{biasing entity}". We used an off-the-shelf speech retriever [5] to retrieve top-1 entity mention from the speech, and replace {biasing entity} with it in the prompt above.…”
Section: Speech Translationmentioning
confidence: 99%
“…In previous work, a joint Speech Language Model (SLM) [5] was introduced using an adapter-based approach [6] to unify pretrained speech and text models for an end-to-end English dialog understanding task, namely, MultiWoz [7]. In this work, we refine the proposed SLM using multilingual speech Fig.…”
Section: Introductionmentioning
confidence: 99%