MOSS-TTSD - 文本到语音对话生成 支持零样本多人语音克隆 支持50系显卡 一键整合包下载

无言***

MOSS-TTSD - 文本到语音对话生成 支持零样本多人语音克隆 支持50系显卡 一键整合包下载 ... ... ... ... .. ...

MOSS-TTSD 是一种语音对话生成模型，能够支持中英文的富有表现力的对话语音合成，支持零样本多说话人语音克隆、语音事件控制和长文本语音生成。
MOSS-TTSD（文本到语音对话）支持中英文的双语语音对话合成，能够将两个说话者之间的对话脚本转化为自然、富有表现力的会话式语音。MOSS-TTSD 支持语音克隆和长单次会话语音生成，非常适合 AI 播客制作。

通俗的讲，只需要上传两段参考音频，输入两个人的对话文字内容，即可一键生成两人的对话音频内容，生成对话语音自然、且富有表现力，几乎接近真人对话水平。


应用领域
AI 播客制作： 自动将文本内容转化为多人对话的播客音频，支持零样本人声克隆和长语音生成，降低制作成本。
影视配音与动画： 为影视作品、动画生成自然对话语音，支持多语言切换，提升配音效率。
长篇访谈与会议记录： 将访谈或会议文本转化为语音，保留对话的韵律和情感，便于存档和分享。
数字人对话带货： 为虚拟主播或数字人提供自然流畅的对话语音，增强互动体验。
语音助手与客服： 提升语音助手的对话自然度，支持多轮对话和情感表达，改善用户体验。


使用教程：（建议N卡，显存8G起。支持50系显卡，基于CUDA12.8）

输入两个人对话文本内容，用标签[S1]和[S2]区分，上传两段说话人的参考音频，生成即可。


2025年7月10日，2楼报错补丁：下载帖子内附件文件 👇，复制到一键包目录下，替换即可。
当前一键包基于40系显卡（sm_90）编译安装FlashAttn，其他显卡请自行下载附件补丁，替换即可。


下载地址：
123云盘：https://www.123684.com/s/OYeA-sO1Bh 提取码:7788
百度网盘：

游客，上上宾会员可免费下载该资源，点此开通上上宾 免费下载全站99%的付费资源。或单独支付 66碎银 下载该资源立即购买

解压密码：https://deepfaces.cc/ 复制这个完整的网址即是解压密码，不要有空格，复制粘贴即可

来日***

下午试了下，音色相似度不高，其他都不错

touv***

Exception raised from c10_cuda_check_implementation at C:\actions-runner\_work\pytorch\pytorch\pytorch\c10\cuda\CUDAException.cpp:43 (most recent call first):
00007FFDBA072C2400007FFDBA072B80 c10.dll!c10::Error::Error [<unknown file> @ <unknown line number>]
00007FFDBA07170A00007FFDBA0716B0 c10.dll!c10::detail::torchCheckFail [<unknown file> @ <unknown line number>]
00007FFDB73F6BBF00007FFDB73F68B0 c10_cuda.dll!c10::cuda::c10_cuda_check_implementation [<unknown file> @ <unknown line number>]
00007FFC5B8E9D8300007FFC5B7EC460 flash_attn_2_cuda.cp310-win_amd64.pyd!PyInit_flash_attn_2_cuda [<unknown file> @ <unknown line number>]
00007FFC5B8E97BE00007FFC5B7EC460 flash_attn_2_cuda.cp310-win_amd64.pyd!PyInit_flash_attn_2_cuda [<unknown file> @ <unknown line number>]
00007FFC5B8E964800007FFC5B7EC460 flash_attn_2_cuda.cp310-win_amd64.pyd!PyInit_flash_attn_2_cuda [<unknown file> @ <unknown line number>]
00007FFC5B7E457200007FFC5B7D5F10 flash_attn_2_cuda.cp310-win_amd64.pyd!c10::ivalue::Object:perator= [<unknown file> @ <unknown line number>]
00007FFC5B7F2C2100007FFC5B7EC460 flash_attn_2_cuda.cp310-win_amd64.pyd!PyInit_flash_attn_2_cuda [<unknown file> @ <unknown line number>]
00007FFC5B7F2D3400007FFC5B7EC460 flash_attn_2_cuda.cp310-win_amd64.pyd!PyInit_flash_attn_2_cuda [<unknown file> @ <unknown line number>]
00007FFC5B7DE92800007FFC5B7D5F10 flash_attn_2_cuda.cp310-win_amd64.pyd!c10::ivalue::Object:perator= [<unknown file> @ <unknown line number>]
00007FFDC1AF4D5D00007FFDC1AF3F90 python310.dll!PyCFunction_GetFlags [<unknown file> @ <unknown line number>]
00007FFDC1AAEFEB00007FFDC1AAEEB0 python310.dll!PyObject_MakeTpCall [<unknown file> @ <unknown line number>]
00007FFDC1BBA81200007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB6CA900007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFDC1BBAA2400007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB59C000007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFDC1BBAA2400007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB59C000007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFDC1BBAA2400007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB59C000007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFDC1BBAA2400007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB59C000007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFC8A8CF89300007FFC8A8A96A0 torch_python.dll!torch::FunctionParameter::type_name [<unknown file> @ <unknown line number>]
00007FFC8A8CB84A00007FFC8A8A96A0 torch_python.dll!torch::FunctionParameter::type_name [<unknown file> @ <unknown line number>]
00007FFC8A8C2E7400007FFC8A8A96A0 torch_python.dll!torch::FunctionParameter::type_name [<unknown file> @ <unknown line number>]
00007FFC8A8A9D6600007FFC8A8A96A0 torch_python.dll!torch::FunctionParameter::type_name [<unknown file> @ <unknown line number>]
00007FFC89FF18D400007FFC89FF0A30 torch_python.dll!c10::ivalue::Future::devices [<unknown file> @ <unknown line number>]
00007FFDC1AF4D5D00007FFDC1AF3F90 python310.dll!PyCFunction_GetFlags [<unknown file> @ <unknown line number>]
00007FFDC1AAEFEB00007FFDC1AAEEB0 python310.dll!PyObject_MakeTpCall [<unknown file> @ <unknown line number>]
00007FFDC1AB18EC00007FFDC1AB1790 python310.dll!PyMethod_Self [<unknown file> @ <unknown line number>]
00007FFDC1BBAA2400007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB59C000007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFDC1AB169100007FFDC1AB12A0 python310.dll!PyCell_Set [<unknown file> @ <unknown line number>]
00007FFDC1AB18EC00007FFDC1AB1790 python310.dll!PyMethod_Self [<unknown file> @ <unknown line number>]
00007FFDC1BBAA2400007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB59C000007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFDC1BBAA2400007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]
00007FFDC1BB59C000007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1BB8E6B00007FFDC1BB3970 python310.dll!PyEval_EvalFrameDefault [<unknown file> @ <unknown line number>]
00007FFDC1AAF3AE00007FFDC1AAF370 python310.dll!PyFunction_Vectorcall [<unknown file> @ <unknown line number>]
00007FFC8A8CF81500007FFC8A8A96A0 torch_python.dll!torch::FunctionParameter::type_name [<unknown file> @ <unknown line number>]
00007FFCE6410B6100007FFCE64109A0 torch_cpu.dll!c10:ispatcher::callBoxed [<unknown file> @ <unknown line number>]
00007FFC8A6A234400007FFC8A6A2280 torch_python.dll!torch::jit::invokeOperatorFromPython [<unknown file> @ <unknown line number>]
00007FFC8A69F45700007FFC8A69F320 torch_python.dll!torch::jit::_get_operation_for_overload_or_packet [<unknown file> @ <unknown line number>]
00007FFC8A60834800007FFC8A5F3D40 torch_python.dll!c10d:ythonOnCompletionHook:ythonOnCompletionHook [<unknown file> @ <unknown line number>]
00007FFC8A5AD8F600007FFC8A535360 torch_python.dll!THPPointer<THPStorage>::dup [<unknown file> @ <unknown line number>]
00007FFC89FF18D400007FFC89FF0A30 torch_python.dll!c10::ivalue::Future::devices [<unknown file> @ <unknown line number>]
00007FFDC1AF4D5D00007FFDC1AF3F90 python310.dll!PyCFunction_GetFlags [<unknown file> @ <unknown line number>]
00007FFDC1AAF2A700007FFDC1AAF1B0 python310.dll!PyObject_Call [<unknown file> @ <unknown line number>]
00007FFDC1BBAB1F00007FFDC1BBA410 python310.dll!PyEval_GetFuncDesc [<unknown file> @ <unknown line number>]

无言***

touvidia 发表于 2025-7-10 11:09
Exception raised from c10_cuda_check_implementation at C:\actions-runner\_work\pytorch\pytorch\pytor ...

Flash_attn跟硬件不兼容导致的
你要有动手能力，手动编译安装Flash_attn
否则等我发个补丁，禁用Flash_attn，推理速度会慢一些

metal***

不支持2080ti,
RuntimeError: FlashAttention only supports Ampere GPUs or newer.

无言***

metalcslug 发表于 2025-7-10 22:26
不支持2080ti,
RuntimeError: FlashAttention only supports Ampere GPUs or newer.

20系不支持半精度，需要改代码。
建议换卡，20系以后会慢慢被放弃