Whispering LLaMA: A Cross-Modal Generative Error Correction Framework for Speech Recognition

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Radhakrishnan, Srijith, Yang, Chao-Han Huck, Khan, Sumeer Ahmad, Kumar, Rohit, Kiani, Narsis A., Gomez-Cabrero, David, Tegner, Jesper N.
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911295495733248
author Radhakrishnan, Srijith
Yang, Chao-Han Huck
Khan, Sumeer Ahmad
Kumar, Rohit
Kiani, Narsis A.
Gomez-Cabrero, David
Tegner, Jesper N.
author_facet Radhakrishnan, Srijith
Yang, Chao-Han Huck
Khan, Sumeer Ahmad
Kumar, Rohit
Kiani, Narsis A.
Gomez-Cabrero, David
Tegner, Jesper N.
contents We introduce a new cross-modal fusion technique designed for generative error correction in automatic speech recognition (ASR). Our methodology leverages both acoustic information and external linguistic representations to generate accurate speech transcription contexts. This marks a step towards a fresh paradigm in generative error correction within the realm of n-best hypotheses. Unlike the existing ranking-based rescoring methods, our approach adeptly uses distinct initialization techniques and parameter-efficient algorithms to boost ASR performance derived from pre-trained speech and text models. Through evaluation across diverse ASR datasets, we evaluate the stability and reproducibility of our fusion technique, demonstrating its improved word error rate relative (WERR) performance in comparison to n-best hypotheses by relatively 37.66%. To encourage future research, we have made our code and pre-trained models open source at https://github.com/Srijith-rkr/Whispering-LLaMA.
format Preprint
id arxiv_https___arxiv_org_abs_2310_06434
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Whispering LLaMA: A Cross-Modal Generative Error Correction Framework for Speech Recognition
Radhakrishnan, Srijith
Yang, Chao-Han Huck
Khan, Sumeer Ahmad
Kumar, Rohit
Kiani, Narsis A.
Gomez-Cabrero, David
Tegner, Jesper N.
Computation and Language
Artificial Intelligence
Multimedia
Sound
Audio and Speech Processing
We introduce a new cross-modal fusion technique designed for generative error correction in automatic speech recognition (ASR). Our methodology leverages both acoustic information and external linguistic representations to generate accurate speech transcription contexts. This marks a step towards a fresh paradigm in generative error correction within the realm of n-best hypotheses. Unlike the existing ranking-based rescoring methods, our approach adeptly uses distinct initialization techniques and parameter-efficient algorithms to boost ASR performance derived from pre-trained speech and text models. Through evaluation across diverse ASR datasets, we evaluate the stability and reproducibility of our fusion technique, demonstrating its improved word error rate relative (WERR) performance in comparison to n-best hypotheses by relatively 37.66%. To encourage future research, we have made our code and pre-trained models open source at https://github.com/Srijith-rkr/Whispering-LLaMA.
title Whispering LLaMA: A Cross-Modal Generative Error Correction Framework for Speech Recognition
topic Computation and Language
Artificial Intelligence
Multimedia
Sound
Audio and Speech Processing
url https://arxiv.org/abs/2310.06434