Views
No views yet
| Model | Type | Download |
|---|---|---|
| GLM-4-Voice-Tokenizer | Speech Tokenizer | 🤗 Huggingface 🤖 ModelScope |
| GLM-4-Voice-9B | Chat Model | 🤗 Huggingface 🤖 ModelScope |
| GLM-4-Voice-Decoder | Speech Decoder | 🤗 Huggingface 🤖 ModelScope |
1git clone --recurse-submodules https://github.com/THUDM/GLM-4-Voice
2cd GLM-4-Voicezhipuai/glm-4-voice:0.1 以跳过这一步。pip install -r requirements.txttransformers 初始化，因此 checkpoint 需要单独下载。1# git 模型下载，请确保已安装 git-lfs
2git lfs install
3git clone https://huggingface.co/THUDM/glm-4-voice-decoderpython model_server.py --host localhost --model-path THUDM/glm-4-voice-9b --port 10000 --dtype bfloat16 --device cuda:0python model_server.py --host localhost --model-path THUDM/glm-4-voice-9b --port 10000 --dtype int4 --device cuda:0glm-4-voice-9b。如果网络条件不好，也手动下载之后通过 --model-path 指定本地的路径。python web_demo.py --tokenizer-path THUDM/glm-4-voice-tokenizer --model-path THUDM/glm-4-voice-9b --flow-path ./glm-4-voice-decoderglm-4-voice-tokenizer 和 glm-4-voice-9b。 请注意，glm-4-voice-decoder 需要手动下载。--tokenizer-path, --flow-path 和 --model-path 指定本地的路径。@misc{zeng2024glm4,
title={GLM-4-Voice: Towards Intelligent and Human-Like End-to-End Spoken Chatbot},
author={Aohan Zeng and Zhengxiao Du and Mingdao Liu and Kedong Wang and Shengmin Jiang and Lei Zhao and Yuxiao Dong and Jie Tang},
year={2024},
eprint={2412.02612},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2412.02612},
}@misc{zeng2024scaling,
title={Scaling Speech-Text Pre-training with Synthetic Interleaved Data},
author={Aohan Zeng and Zhengxiao Du and Mingdao Liu and Lei Zhang and Shengmin Jiang and Yuxiao Dong and Jie Tang},
year={2024},
eprint={2411.17607},
archivePrefix={arXiv},
primaryClass={cs.CL},
url={https://arxiv.org/abs/2411.17607},
}