Abstractive summarization from Audio Transcription

TL;DR

This paper presents an end-to-end audio summarization model that utilizes fine-tuning techniques like LoRA and quantization to make large language models more accessible for audio transcription summarization tasks.

Contribution

It introduces a novel E2E audio summarization approach employing fine-tuning methods to reduce resource requirements and evaluates their effectiveness.

Findings

Fine-tuning methods improve model efficiency

Techniques are applicable to audio summarization tasks

Resource reduction enables broader use of large models

Abstract

Currently, large language models are gaining popularity, their achievements are used in many areas, ranging from text translation to generating answers to queries. However, the main problem with these new machine learning algorithms is that training such models requires large computing resources that only large IT companies have. To avoid this problem, a number of methods (LoRA, quantization) have been proposed so that existing models can be effectively fine-tuned for specific tasks. In this paper, we propose an E2E (end to end) audio summarization model using these techniques. In addition, this paper examines the effectiveness of these approaches to the problem under consideration and draws conclusions about the applicability of these methods.