DiarizationLM: Speaker Diarization Post-Processing with Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Quan, Huang, Yiling, Zhao, Guanlong, Clark, Evan, Xia, Wei, Liao, Hank
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910777296814080
author Wang, Quan
Huang, Yiling
Zhao, Guanlong
Clark, Evan
Xia, Wei
Liao, Hank
author_facet Wang, Quan
Huang, Yiling
Zhao, Guanlong
Clark, Evan
Xia, Wei
Liao, Hank
contents In this paper, we introduce DiarizationLM, a framework to leverage large language models (LLM) to post-process the outputs from a speaker diarization system. Various goals can be achieved with the proposed framework, such as improving the readability of the diarized transcript, or reducing the word diarization error rate (WDER). In this framework, the outputs of the automatic speech recognition (ASR) and speaker diarization systems are represented as a compact textual format, which is included in the prompt to an optionally finetuned LLM. The outputs of the LLM can be used as the refined diarization results with the desired enhancement. As a post-processing step, this framework can be easily applied to any off-the-shelf ASR and speaker diarization systems without retraining existing components. Our experiments show that a finetuned PaLM 2-S model can reduce the WDER by rel. 55.5% on the Fisher telephone conversation dataset, and rel. 44.9% on the Callhome English dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2401_03506
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DiarizationLM: Speaker Diarization Post-Processing with Large Language Models
Wang, Quan
Huang, Yiling
Zhao, Guanlong
Clark, Evan
Xia, Wei
Liao, Hank
Audio and Speech Processing
Machine Learning
Sound
In this paper, we introduce DiarizationLM, a framework to leverage large language models (LLM) to post-process the outputs from a speaker diarization system. Various goals can be achieved with the proposed framework, such as improving the readability of the diarized transcript, or reducing the word diarization error rate (WDER). In this framework, the outputs of the automatic speech recognition (ASR) and speaker diarization systems are represented as a compact textual format, which is included in the prompt to an optionally finetuned LLM. The outputs of the LLM can be used as the refined diarization results with the desired enhancement. As a post-processing step, this framework can be easily applied to any off-the-shelf ASR and speaker diarization systems without retraining existing components. Our experiments show that a finetuned PaLM 2-S model can reduce the WDER by rel. 55.5% on the Fisher telephone conversation dataset, and rel. 44.9% on the Callhome English dataset.
title DiarizationLM: Speaker Diarization Post-Processing with Large Language Models
topic Audio and Speech Processing
Machine Learning
Sound
url https://arxiv.org/abs/2401.03506