Publication detail
Speculative Speech Recognition by Audio-Prefixed Low-Rank Adaptation of Language Models
YUSUF, B. BASKAR, M. ROSENBERG, A. RAMABHADRAN, B.
Original Title
Speculative Speech Recognition by Audio-Prefixed Low-Rank Adaptation of Language Models
Type
conference paper
Language
English
Original Abstract
This paper explores speculative speech recognition (SSR), where we empower conventional automatic speech recognition (ASR) with speculation capabilities, allowing the recognizer to run ahead of audio. We introduce a metric for measuring SSR performance and we propose a model which does SSR by com bining a RNN-Transducer-based ASR system with an audioprefixed language model (LM). The ASR system transcribes ongoing audio and feeds the resulting transcripts, along with an audiodependent prefix, to the LM, which speculates likely completions for the transcriptions. We experiment with a variety of ASR datasets on which show the efficacy our method and the feasibility of SSR as a method of reducing ASR latency.
Keywords
low-latency speech recognition, speculative speech recognition, prefix language model, low-rank adaptation
Authors
YUSUF, B.; BASKAR, M.; ROSENBERG, A.; RAMABHADRAN, B.
Released
1. 9. 2024
Publisher
International Speech Communication Association
Location
Kos
ISBN
1990-9772
Periodical
Proceedings of Interspeech
Year of study
2024
Number
9
State
French Republic
Pages from
792
Pages to
796
Pages count
5
URL
BibTex
@inproceedings{BUT193739,
author="YUSUF, B. and BASKAR, M. and ROSENBERG, A. and RAMABHADRAN, B.",
title="Speculative Speech Recognition by Audio-Prefixed Low-Rank Adaptation of Language Models",
booktitle="Proceedings of Interspeech 2024",
year="2024",
journal="Proceedings of Interspeech",
volume="2024",
number="9",
pages="792--796",
publisher="International Speech Communication Association",
address="Kos",
doi="10.21437/Interspeech.2024-298",
issn="1990-9772",
url="https://www.isca-archive.org/interspeech_2024/yusuf24_interspeech.pdf"
}
Documents