{"id":13407457,"url":"https://github.com/LinkSoul-AI/LLaSM","last_synced_at":"2025-03-14T12:31:01.898Z","repository":{"id":184996362,"uuid":"672784891","full_name":"LinkSoul-AI/LLaSM","owner":"LinkSoul-AI","description":"第一个支持中英文双语语音-文本多模态对话的开源可商用对话模型。便捷的语音输入将大幅改善以文本为输入的大模型的使用体验，同时避免了基于 ASR 解决方案的繁琐流程以及可能引入的错误。","archived":false,"fork":false,"pushed_at":"2023-09-11T05:29:06.000Z","size":3415,"stargazers_count":529,"open_issues_count":5,"forks_count":54,"subscribers_count":14,"default_branch":"main","last_synced_at":"2024-10-18T21:17:20.972Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"https://github.com/LinkSoul-AI/LLaSM","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/LinkSoul-AI.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2023-07-31T06:55:36.000Z","updated_at":"2024-10-17T08:04:15.000Z","dependencies_parsed_at":null,"dependency_job_id":"6b52d11e-0bca-491d-a672-da6a007fa9ae","html_url":"https://github.com/LinkSoul-AI/LLaSM","commit_stats":null,"previous_names":["linksoul-ai/llasm"],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LinkSoul-AI%2FLLaSM","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LinkSoul-AI%2FLLaSM/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LinkSoul-AI%2FLLaSM/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/LinkSoul-AI%2FLLaSM/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/LinkSoul-AI","download_url":"https://codeload.github.com/LinkSoul-AI/LLaSM/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":243495506,"owners_count":20299922,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-07-30T20:00:40.709Z","updated_at":"2025-03-14T12:31:01.887Z","avatar_url":"https://github.com/LinkSoul-AI.png","language":"Python","funding_links":[],"categories":["Multimodal LLM","Project List","多模态大模型"],"sub_categories":["\u003cspan id=\"tool\"\u003eLLM (LLM \u0026 Tool)\u003c/span\u003e","网络服务_其他"],"readme":"# LLaSM: Large Language and Speech Model\n\n[![](https://img.shields.io/badge/LLaSM-Chinese-blue)](https://huggingface.co/spaces/LinkSoul/LLaSM) [![](https://img.shields.io/badge/Commercial-Support-blue)](https://huggingface.co/spaces/LinkSoul/LLaSM) [![](https://img.shields.io/badge/License-Apache_v2-blue)](https://github.com/LinkSoul-AI/LLaSM/blob/main/LICENSE) [![](https://img.shields.io/badge/Paper-arXiv-red)](https://arxiv.org/abs/2308.15930) [![](https://img.shields.io/badge/HuggingFace-Live_Demo-green)](https://huggingface.co/spaces/LinkSoul/LLaSM) [![](https://img.shields.io/badge/Datasets-LLaSM_Audio_Instructions-yellow)](https://huggingface.co/datasets/LinkSoul/LLaSM-Audio-Instructions)\n\n开源，可商用的**中英文双语语音-语言助手 LLaSM 以及中英文语音 SFT 数据集 LLaSM-Audio-Instructions**，第一个支持中英文语音-文本多模态对话的开源可商用对话模型。\n\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\".github/preview.jpg\" width=\"40%\"\u003e\n\u003c/p\u003e\n\n## 模型框架\n\n![Framework](.github/framework.png)\n\n## 基础演示\n\n![Base Demo](.github/demo.gif)\n\n## 在线试玩\n\n\u003e Talk is cheap, Show you the Demo.\n- [Demo 地址 / Hugging Face Spaces](https://huggingface.co/spaces/LinkSoul/LLaSM) \n\n## 论文\n- arXiv 链接：https://arxiv.org/abs/2308.15930\n\n## 资源下载\n\n- Hugging Face模型下载：\n  - [LLaSM-Chinese-Llama-2-7B](https://huggingface.co/LinkSoul/LLaSM-Cllama2)\n  - [LLaSM-Baichuan-7B](https://huggingface.co/LinkSoul/LLaSM-Baichuan)\n\n- 百度网盘下载:\n  - [LLaSM-Chinese-Llama-2-7B](https://pan.baidu.com/s/1PaipNDfqV7f3W1-tl5rwzA?pwd=2549)\n  - [LLaSM-Baichuan-7B](https://pan.baidu.com/s/1QZrXA8IJXclN77T4jM7tEw?pwd=y2p7)\n\n- 语言模型:\n  - [Chinese-Llama-2-7b](https://github.com/LinkSoul-AI/Chinese-Llama-2-7b)\n  - [Baichuan-7B](https://huggingface.co/baichuan-inc/Baichuan-7B)\n\n- 数据集：[LLaSM-Audio-Instructions](https://huggingface.co/datasets/LinkSoul/LLaSM-Audio-Instructions)\n\n## 环境安装\n```shell\n# clone the repository\ngit clone https://github.com/LinkSoul-AI/LLaSM\ncd LLaSM\n\n# install package\nconda create -n llasm python=3.10 -y\nconda activate llasm\npip install --upgrade pip\npip install -e .\n```\n\n## 快速测试\n- 下载 Whisper large v2 模型：https://huggingface.co/openai/whisper-large-v2\n\n```shell\nexport LLASM_DEVICE=\"cuda:0\"\npython infer.py \\\n    --input_audio_file PATH/TO/YOUR/AUDIO \\\n    --llasm_model PATH/TO/LLaSM/MODEL \\\n    --llasm_audio_tower PATH/TO/WHISPER/MODEL \\\n    --llm_type \"Chinese_llama2\" or \"baichuan\" \\\n```\n\n## TODO\n- 如何训练\n- int4 量化\n- docker 部署\n\n## 相关项目\n- [Chinese-Llama-2-7B](https://huggingface.co/LinkSoul/Chinese-Llama-2-7b)\n- [Whisper](https://ai.meta.com/llama/)\n- [baichuan-inc/Baichuan-7B](https://huggingface.co/baichuan-inc/Baichuan-7B)\n\n\n## 项目协议\n\n[Apache-2.0 license](https://github.com/LinkSoul-AI/LLaSM/blob/main/LICENSE)\n\n## Citation\n\n如果您发现我们的工作和此仓库有用，欢迎给一个星星 :star: 鼓励我们一下 :beer::\n```bibtex\n@misc{shu2023llasm,\n      title={LLaSM: Large Language and Speech Model}, \n      author={Yu Shu and Siwei Dong and Guangyao Chen and Wenhao Huang and Ruihua Zhang and Daochen Shi and Qiqi Xiang and Yemin Shi},\n      year={2023},\n      eprint={2308.15930},\n      archivePrefix={arXiv},\n      primaryClass={cs.CL}\n}\n```\n\n## 微信交流群\n\n\u003cimg src=\".github/QRcode.jpg\" alt=\"微信交流群\" width=\"300\"/\u003e\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FLinkSoul-AI%2FLLaSM","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FLinkSoul-AI%2FLLaSM","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FLinkSoul-AI%2FLLaSM/lists"}