arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2407.02005cs.CLcs.SDeess.AS
收藏

使用大型语言模型的端到端语音摘要生成

An End-to-End Speech Summarization Using Large Language Model

Huawei Translation Service Center(华为翻译服务中心)

Hengchao Shang, Zongyao Li, Jiaxin Guo, Shaojun Li, Zhiqiang Rao, Yuanchang Luo, Daimeng Wei, Hao Yang

AI总结:

本文提出以 Q-Former 连接语音与 LLM 的端到端语音摘要模型,通过 ASR、TSum 辅助任务和课程学习实现多阶段训练,并在 How-2 上取得有竞争力的结果。

摘要:

抽象式语音摘要生成(SSum)旨在根据语音内容生成类人文本摘要。该任务在处理长语音输入,以及捕捉长语音输入与短文本摘要之间复杂的跨模态映射方面面临困难。有关大型语言模型(LLM)和多模态信息融合的研究为应对这些挑战提供了新的思路。本文提出一种端到端 SSum 模型,该模型使用 Q-Former 作为音频—文本模态之间的连接器,并利用 LLM 直接从语音特征生成文本摘要。我们采用多阶段训练方法,其中包括以基于 LLM 的自动语音识别(ASR)和文本摘要生成(TSum)任务作为辅助任务。ASR 任务用于对齐特征空间,并增强 LLM 处理更长语音的能力。随后,我们采用课程学习策略,促进模型从 TSum 向 SSum 过渡。最终,我们的模型在 How-2 数据集上取得了具有竞争力的性能。

英文摘要:

Abstractive Speech Summarization (SSum) aims to generate human-like text summaries from spoken content. It encounters difficulties in handling long speech input and capturing the intricate cross-modal mapping between long speech inputs and short text summaries. Research on large language models (LLMs) and multimodal information fusion has provided new insights for addressing these challenges. In this paper, we propose an end-to-end SSum model that utilizes Q-Former as a connector for the audio-text modality and employs LLMs to generate text summaries directly from speech features. We adopt a multi-stage training approach that includes LLM based ASR and Text Summarization (TSum) tasks as auxiliary tasks. ASR tasks are used to align feature spaces and enhance the LLM's ability to handle longer speech. Then, we utilize a curriculum learning strategy to facilitate the model's transition from TSum to SSum. Finally, our model achieves competitive performance on the How-2 dataset.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑