search

CCmahua / ChatTTS-Enhanced

Other
475 stars 66 forks source link

尊敬的开发者您好 提交一个测试bug #2

Closed passwordhk closed 3 months ago

passwordhk commented 3 months ago

我的系统环境是 intel 13400F 搭配Amd显卡，在测试环境中发现无法正常自定义音色，有以下报错

Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

请问解决方案是去买个显卡吗 哭死

CCmahua commented 3 months ago

processors/audio_processor.py第38行修改为

 rand_spk = torch.load(emb_upload_path,map_location=torch.device('cpu'))

processors/config_processor.py的save_config函数改为

def save_config(name,con_params: ConfigParams):
    con_params.tensor = get_seed_tensor(con_params.audio_seed)
    if con_params.custom_emb and con_params.emb_path is not None:
        con_params.tensor = torch.load(con_params.emb_path, map_location=torch.device('cpu'))
    save_path = os.path.join(get_path('CONFIG_DIR'), f"{name}.pt")
    torch.save(con_params, save_path)
    return f"配置保存至 {save_path}"
passwordhk commented 3 months ago

已解决问题，十分感谢开发者！