VoxCPM 1.5.0 macOS

Python 版本

python3 --version

Python3.11.13

python3 -c"import torch; print(f'PyTorch: {torch.__version__}'); print(f'MPS Available: {torch.backends.mps.is_available()}'); print(f'MPS Built: {torch.backends.mps.is_built()}')"2>/dev/null||echo"PyTorch 未安装"

PyTorch:2.9.0 MPS Available: True MPS Built: True

python3 -c"import torchaudio; print(f'TorchAudio: {torchaudio.__version__}')"2>/dev/null||echo"TorchAudio 未安装"

TorchAudio:2.9.0

卸载现有版本

pip uninstall voxcpm -y

Found existing installation: voxcpm1.0.5 Uninstalling voxcpm-1.0.5: Successfully uninstalled voxcpm-1.0.5

重新安装

pipinstallvoxcpm

启动 Web

python app.py

🚀 Running on device: cpu funasr version:1.2.7. Downloading Model from https://www.modelscope.cn to directory: /Users/m4/.cache/modelscope/hub/models/iic/SenseVoiceSmall2025-12-2722:47:19,955 - modelscope - WARNING - Using branch: master as version is unstable, use with caution WARNING:root:trust_remote_code: False * Running onlocalURL: http://localhost:7860 * To create a public link,set`share=True`in`launch()`.

下载模型

• 下载到项目 models 目录

python3 -c" from modelscope import snapshot_download model_dir = snapshot_download('OpenBMB/VoxCPM1.5', cache_dir='./models') print(f'Model downloaded to: {model_dir}') "

python3 -c"import transformers; print(f'transformers: {transformers.__version__}')"

transformers:4.57.1

降级 transformers 到兼容版本

pipinstall"transformers>=4.36.2,<4.50"

python3 -c"import transformers; print(f'transformers: {transformers.__version__}')"transformers:4.49.0

降级 NumPy 到 1.x 版本

pipinstall"numpy<2"pipinstall"numpy<2"-i https://pypi.tuna.tsinghua.edu.cn/simple

datasets 库版本

pipinstall"datasets==2.18.0"-i https://pypi.tuna.tsinghua.edu.cn/simple

开始生成