LongSpeech-Eval

# The proposed Long-Speech understanding evaluation dataset for the paper '[FastLongSpeech: Enhancing Large Speech-Language Models for Efficient Long-Speech Processing](https://arxiv.org/abs/2507.14815)'

## Usage

First download the model from [Model](https://huggingface.co/ICTNLP/FastLongSpeech).

Then please refer to [Github Page](https://github.com/ictnlp/FastLongSpeech).

### Requirements

We suggest to run with Python 3.10.
Examples of usage:
```
git clone https://github.com/ictnlp/FastLongSpeech.git
cd transformers-main
pip install -e .
pip install deepspeed sentencepiece librosa
```

## Training Model
- https://huggingface.co/ICTNLP/FastLongSpeech
## Github Pages
- https://github.com/ictnlp/FastLongSpeech

# 对应论文《FastLongSpeech：面向高效长语音处理的大规模语音语言模型优化》（[FastLongSpeech: Enhancing Large Speech-Language Models for Efficient Long-Speech Processing](https://arxiv.org/abs/2507.14815)）所提出的长语音理解评测数据集

## 使用方法
首先从[模型仓库](https://huggingface.co/ICTNLP/FastLongSpeech)下载对应模型。
随后请参考[GitHub 仓库页面](https://github.com/ictnlp/FastLongSpeech)完成后续操作。

### 依赖要求
我们建议使用Python 3.10版本运行本项目。
使用示例如下：

git clone https://github.com/ictnlp/FastLongSpeech.git
cd transformers-main
pip install -e .
pip install deepspeed sentencepiece librosa

## 训练模型
- https://huggingface.co/ICTNLP/FastLongSpeech

## GitHub 仓库页面
- https://github.com/ictnlp/FastLongSpeech

搜集汇总

背景与挑战

背景概述

LongSpeech-Eval是一个用于评估长语音理解的数据集，与论文'FastLongSpeech'相关，旨在提升大语音语言模型在长语音处理中的效率。该数据集由ICTNLP发布，采用Apache-2.0许可证，并于2025年7月22日更新。